TAX AGGRESSIVENESS, CORPORATE GOVERNANCE, AND FIRM VALUE: AN EMPIRICAL EVIDENCE FROM THAILAND RAWIWAN KOANANTACHAI

Transcription

1 TAX AGGRESSIVENESS, CORPORATE GOVERNANCE, AND FIRM VALUE: AN EMPIRICAL EVIDENCE FROM THAILAND RAWIWAN KOANANTACHAI MASTER OF SCIENCE PROGRAM IN FINANCE (INTERNATIONAL PROGRAM) FACULTY OF COMMERCE AND ACCOUNTANCY THAMMASAT UNIVERSITY, BANGKOK, THAILAND MAY 2013

2 TAX AGGRESSIVENESS, CORPORATE GOVERNANCE, AND FIRM VALUE: AN EMPIRICAL EVIDENCE FROM THAILAND RAWIWAN KOANANTACHAI MASTER OF SCIENCE PROGRAM IN FINANCE (INTERNATIONAL PROGRAM) FACULTY OF COMMERCE AND ACCOUNTANCY THAMMASAT UNIVERSITY, BANGKOK, THAILAND MAY 2013

3 Tax Aggressiveness, Corporate Governance, and Firm Value: An Empirical Evidence from Thailand Rawiwan Koanantachai An Independent Study Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science (Finance) Master of Science Program in Finance (International Program) Faculty of Commerce and Accountancy Thammasat University, Bangkok, Thailand May 2013

4 Thammasat University Faculty of Commerce and Accountancy An Independent Study By Rawiwan Koanantachai Tax Aggressiveness, Corporate Governance, and Firm Value: An Empirical Evidence from Thailand has been approved as a partial fulfillment of the requirements for the Degree of Master of Science (Finance) On May, 2013 Advisor: (Ajarn Dr. Chaiyuth Padungsaksawasdi) Co-Advisor: (Assoc. Prof. Dr. Pornsit Jiraporn)

5 Tax Aggressiveness, Corporate Governance, and Firm Value: An Empirical Evidence from Thailand ABSTRACT This paper investigates the relationship between corporate governance affects tax aggressiveness in the Stock Exchange of Thailand (SET) during Further, I study the impact of these two variables on the value of the firms. The results show that corporate governance is positively related to tax avoidance, meaning that firms with good governance pay tax less than firms with bad governance. The board of directors and audit committee play an important role on tax reduction. Interestingly, foreign holding of the company helps reduce tax payment. Moreover, firms with better governance lead to good performance on firm valuation. 1

6 INTRODUCTION Income tax expense is one of the important costs of a corporation. The corporate tax rate on Southeast Asia countries is around 25% to 35% of earnings before tax. Generally, tax rates are decreased over the , increasing competitive advantage in international environment. Figure 1 depicts the company corporate tax rate on Southeast Asia countries, normally Indonesia, Malaysia, the Philippines, Singapore and Thailand. To achieve the goal of maximization wealth, a company is to decrease tax payment. Corporate income tax rate (%) Figure1: The comparable corporate income tax rate in Southeast Asia during Source: Philippines Thailand Malaysia Singapore Indonesia Tax aggressiveness (tax avoidance) is a legal treatment on tax reduction. Alternatively, it takes advantage of all legal opportunities to minimize income tax, for example investing in another limited company under Thai laws; the dividend received is free of taxes (Revenue Code, section 65 bis (10)). 1 The current and significant issue in Southeast Asia today is the creation of the ASEAN Economic Community (AEC) by 2015, giving opportunities to local business into international context especially in the ASEAN. The AEC blueprint provides that the member countries shall work towards establishing an effective network of bilateral agreements on avoidance of double taxation. 1 Tax avoidance is different from tax evasion, which is the employment of unlawful methods to elude the payment of taxes. Tax evasion is a crime; however, tax avoidance is not. 2

7 Not only eliminate tariffs on the flow of goods within the region, but also there is the issue that concerns tax policies of the member countries. Moreover, the AEC harmonizes national taxation policies, so that member can compete in related with investment and fund flow within ASEAN. Thailand is to reduce the corporate tax rate before the AEC in Thailand s corporate tax rate decreases from 30 percent to 23 percent in 2012 and ultimately to 20 percent. Thailand is considered as one of the AEC members to reduce corporate tax rate. From the importance of corporate income tax as discussed above, tax planning can use as a tool to manage income according to the scandal of Enron in Enron avoided paying corporate income tax for four years before going bankruptcy, although the financial statements were profitable. Enron used tax benefit by stock option, a tax deduction that did not present in the report to manage earnings. Since the failure of Enron, the Sarbanes-Oxley Act was approved in 2002 to make sure that business managers have no excuses for ignorance about their tax function. The regulation covers issues such as auditor independence, corporate governance, internal control assessment, and enhanced financial disclosure (PricewaterhouseCoopers, 2008). Since then, corporate governance calls for attention around the world. The objective of this study is to investigate the relation between tax aggressiveness and corporate governance. Later, I study the impact of these two variables on the value of the firm. For the purpose of the study, I employ effective tax rate as a measure of aggressive tax and create corporate governance index to present corporate governance, as well as calculate for the firm value. The following of this paper is followed. Section I discusses relevant literature. Section II describes theoretical framework related to tax aggressiveness and corporate governance. Section III is data description and methodology. Section IV discusses hypotheses and the role of control variables. Section V presents empirical. The last section is conclusion. 3

8 I. LITERATURE REVIEW Not many study the linkage among tax aggressiveness, corporate governance, and firm. Some shows that tax management enhances earnings, subsequently increasing the firm value (Chen, Dhaliwal, and Trombley, 2007). Minnick and Noga (2010) examine how corporate governance plays a role in long-run tax management. They show that tax avoidance or tax aggressiveness makes benefit to shareholders and gives higher return by transferring wealth from state to the firm s shareholders. Aggressive tax is possible due to tax planning, presenting in the financial statements through the book-tax differences. Book-tax differences emerge from a variety of sources and their result to have managed earnings, related directly with earnings properties and with market participant s valuation such as lower persistence associated with credit ratings (Hanlon and Heitsman, 2010). Tax avoidance logically reflects the firm s cost of debt. Lim (2010) examines the impact of tax avoidance on the cost of debt and its interaction with shareholder activism. The result shows that there is a negative relation between tax avoidance and the cost of debt, supporting the trade-off theory. Through the past few years, there are many new tax technologies or new applications to apply with the firms. Thus tax authorities are more and more aggressive around the world and having been done their best to make sure that corporations can have no excuses for ignorance or evasion about their tax function. And other regulators such as accounting standards organizations, with rules as the Sarbanes- Oxley Act have ensured that the relationship between taxes has taken on an importance for the rest of the business (PricewaterhouseCoopers, 2008). However, there are some researches to indicate that tax avoidance can create the hidden agenda such as hide bad new or mislead investors. Kim and Zhang (2011) study the relation of future stock crash risk and tax avoidance and also further investigate the association or change between tax avoidance and crash risk when external monitoring is effective. The research shows that there are strong evidences to state that tax avoidance is positively associated with the future crash risk. Nevertheless, this crash risk could be reduced when firms have strong external monitoring mechanism 4

9 such as high institutional ownership, high analyst coverage and greater takeover threat from corporate control market. Discrepancy in details of ownership structure, Shleifer and Vishny (1997) indicate that ownership structure plays a significant role for firm creation and performance. Villalonga and Amit (2006) study the effect of firm value from family ownership and control. They find that in case of certain form of family and management control, family ownership will create value. In contrast, when family control is excess, it reduces firm value by the reduction in value being proportional to the excess of voting over cash flow rights. For tax purpose, the research of Huizinga and Nicodème (2003) find that foreign ownership has significant impact to tax planning. Increase of foreign ownership percentage lead to positive level of increase of corporate income tax. Generally, multinational firm could exempt tax expense from foreign source of income in foreign countries according to avoid double taxation. Therefore, foreign investment is one of tools to manage tax of the company in order to take this tax advantage to operate in high-tax countries. Then a higher tax country will attract additional multination firms for more investment. There is literature to research the relation between family firms and tax aggressive. Desai and Dharmalapa (2006), state that tax aggressiveness activities are often bundled with rent extraction. If tax management is used for rent extraction, family owners are willing to save tax. While, there is some opposite research. Chen, Chen, Cheng, and Shevlin (2010), show the result that family firms avoid tax less than non-family firms because they concern more in potential penalty and reputation. Generally Thai firms is set up from family firms, therefore, there is unclear whether family firms will be more of less tax avoidance according to the benefit of tax aggressiveness and cost, which will move higher too. Board of directors or managements will directly affect to tax planning. Dyreng, Hanlon and, Maydew (2010) whom research question are whether individual executives matter for firm tax avoidance. They conclude that individual executives play a significant role in determining the level of tax avoidance that firm undertake. Management and the role of board of directors are the one significant factor to corporate governance (Adams, Hermalin, and Weisbach, 2010). Hermalin and 5

10 Weisbach (2003) point out that board of directors are an integral part of the governance of large organizations, including all corporate and many non-corporate firms from their study over board of directors characteristics and effect of them. For the relation between corporate governance and level of tax aggressiveness, Desai and Dharmapala (2007) perform further investigation of the degree to which corporate tax avoidance activity is valued by investors. The paper says that higher quality firm governance leads to a larger effect of tax avoidance of firm value. Wang (2010) summarizes that transparent firms avoid more tax relative to their counterparts. Corporate transparency will facilitate the monitoring of managerial actions and cite to outside investors to concern the hidden point related with tax avoidance. 6

11 II. THEORITICAL FRAMEWORK a. Tax aggressiveness Since the difference between book accounting and tax accounting deducts taxable income from firm s profit, which book accounting is used on a company s audited financial statements under the accounting standards (GAAP) while tax accounting is income and deductions reported on tax submission in accordance with the authorities rules such as Revenue Code, this difference can be increased by either (1) the opportunistic increase of financial income (earnings management) or (2) the intentional decrease of taxable income (tax avoidance) (Lim, 2010). Tax aggressiveness or tax avoidance is defined as the legal utilization of the tax regime to own advantage, to reduce the amount of tax that is payable by means that are within the law. Whereas tax evasion, on the other hand, is the general term which is an effort of entities to evade tax with illegal meaning. Aggressive tax activities are traditionally viewed as tax saving method in order to transfer benefit from the state to shareholder. Therefore, these activities could also cause increasing of shareholder s wealth through the rise of firm value (Wang, 2010). While the traditional view of corporate tax avoidance suggests that shareholder value should increase with tax avoidance activity, the alternative view provides a few different sides. The research of Desai and Dharmapala (2007), find that corporate governance should be an important determinant of the valuation of purported corporate tax savings. Meanwhile the direct effect of tax avoidance is to increase the after tax value of the firm, these effects are potentially offset, particularly in poorly-governed firms, by the increased opportunities for managerial rent extraction. 7

12 b. Corporate governance Refer to Jensen and Meckling (1976), agency theory describes the relationship between one party, called the principal delegating work to another, called the agent. The theory explains their differences in behavior or decisions that the two parties often have different goals and, independent of their respective goals, may have different attitudes toward risk. Principals employ agents for maximizing return to shareholders whereas agents do for own interesting, also it will create agency problem. In order to mitigate this issue between principals and agents, corporate governance measurement would be applied. For equity ownership, holding share by managers can reduce agency conflict between ownership and management, accordingly, it helps to align their interest with those of outside shareholders. Foreigner and institutional structure could lead to more monitor firm operation, which the appointment of outside directors can also result in an increase in monitoring activities. As literature, Minnick and Noga (2010) examine four corporate governance indexes from board composition, entrenchment, director compensation and executive compensation in their research. According to Wang 2010, the paper ranks four individual proxies for opacity, related to earnings quality, analyst following, press coverage and trading volume. However, this paper constructs the research in Thailand, an emerging market, which has not much information to support this index, compared with firms in the developed making, U.S. I use corporate governance index based on the research of Prommin et al. (2011) which the paper creates corporate governance index in Thailand to measure as corporate governance in this paper as below. (1) Board of directors: They play a role of internal monitoring to protect shareholder s interest. The board of directors has legal authority to hire or fire and compensate top management also deals with problems. Whereas non-executive directors are independent. More independent board result to improve company s performance and increase shareholder wealth (Minnick and Noga, 2010). 8

13 (2) Audit: Audit Committee and external auditors play a role to provide reasonable assurance on the financial reporting, operating efficiency, and compliance so that shareholder could ensure the firm information in relative views. (3) Compensation: Performance-based compensation is designed to relate some proportion of salary to individual performance. (4) Director nominations: They play a role in overseeing matters of corporate governance for the board, including formulating and recommending governance principles and policies in order to enhance the quality and integrity for the corporation. Another framework is information asymmetric. There is unequally information between parties which one party has better relevant information than another. It will lead to an imbalance of power in transactions. Cai, Liu, and Qian (2009) research the relation between information asymmetry and corporate governance. The paper finds that asymmetric information leads to higher costs of direct board monitoring, one factor of corporate governance used in this study. Therefore firm s asymmetric information environment is an important impact on governance mechanisms. Kanagaretnam, Lobo, and Whalen (2007) also examine the relationship between the quality of corporate governance and information asymmetry in the equity market around quarterly earnings announcements. The results show that firms with higher levels of corporate governance have lower information asymmetry around quarterly earnings announcements. 9

14 III. DATA AND METRODOLOGY My sample consists of the firms listed on the Stock Exchange of Thailand (SET) during As of August 2012, listed firms in SET are 512 firms. Data is excluded firms in financial industry, companies under rehabilitation and also I exclude property fund sector 40 firms. According to Thai regulation, property fund is granted to exempt corporate income tax. Therefore, my data include 1,876 companies in 7 industries. Firm-years for SET as of August 2012, ,560 Less: firm-years under financial industry 280 Less: firm-years under rehabilitation 105 Less: firm-years under property fund sector 200 Less: firm-years with missing tax aggressiveness measure 94 Less: firm-years with missing control variables 5 1,876 The prior research, Wang (2010) presents that directly studied tax avoidance, corporate governance or transparency and firm value is mainly performed by data from U.S. However, according to the difference of regulatory and governance environment, the relation will not be the same in other countries especially in emerging market. For emerging countries, I select Thailand as one of the top three of market capitalization per GDP in these Southeast Asian. Thai market capitalization per GDP in term of USD based on the ended of December 2011 is 79% (Data from Stock Exchange of Thailand, 2012). Furthermore, traditional Thai firms are operated by family structure, however, after the development of Thailand economy, the structure have changed. There are more investors from foreigners, financial institution to join the ownership in the firms through direct investment or indirect investment via holding company. So, the current ownership structure comprises with different types. According to trading data from SET as of September 2012, transaction by investors types are 55%, 24% and 22% for local investor, foreign investor and financial institution respectively, compared with the proportion of trading investors in 2002 which are 72%, 22%, and 4%. 10

15 Data for tax aggressiveness, which are extracted by financial statements and information for companies characteristic, come from SETSMART, DataStream and companies website. For corporate governance, data are obtained from firms annual report, 56-1 and SETSMART. Ownership structure is obtained from major shareholders data from SETSMART. a. Measure of tax aggressiveness Refer to Minnick and Noga (2010), I use the annual GAAP ETR, which the firms disclose this expense in the income statements. The GAAP ETR is computed by the ratio of total tax expense and pre-tax income for a given firm as below: GAAP ETR = Tax Expense / Pre-tax Income According to the limitation of traditional of effective tax rate (GAAP accrual basis), which it could create temporary difference or estimation of tax expense such as delaying of revenue recognition in the financial statements. In addition, under IAS no. 12: Income Taxes, deferred tax represent paid or refunded tax in the future according to temporary of timing book-tax differences. This difference is generally used for tax management tool to reduce current tax and maximize the time value of money. Therefore, GAAP ETR will not reflect short-term tax avoidance. To solve this issues, I apply cash effective tax rate (Cash ETR), extracted from cash tax paid, as the measurement of aggressive tax. From Dyreng, Hanlon, and Maydew (2008), the paper shows that there are at least two advantages to use cash effective tax instead of GAAP effective tax. First cash effective tax rate includes an effect of the tax benefits of employee stock options, whereas GAAP effective tax rate does not. Second, cash effective tax rate includes the effect of change in accounting estimate such as allowance expense or tax contingencies reserve which the GAAP tax rate does not take into account. 11

16 Annual cash effective tax rate computes from tax payment, which is presented in the statement of cash flows, divided by income before tax less nonrecurring item as below. Cash ETR = Cash Tax Paid / (Pre-tax Income Special Item) b. Measure of corporate governance From the paper of Prommin et al. (2011), corporate governance index is generated by nine governance standards comprised with the four factors of board of directors, audit, compensation, and director nominations as the table I. I award zero-point in each of factor when no governance standard, one-point when the corporations meet the standard and two-point when the firms practice is above standard. Then the governance index summarizes from zero to eighteen, which zero is the lowest governance quality and eighteen is the highest. Table I: Corporate Governance Index Corporate governance index is constructed based on the governance standard listed in the table. The index comprises with four category; board of directors, audit, compensation, and director nomination, decomposing with nine standard. Governance index is awarded when standard is satisfied in each category. Therefore, corporate governance index would range from zero to eighteen. Category Governance Standard Governance index Board of directors Independent directors Chairman and CEO are separate Largest director s shareholding percentage of issued capital (as measured by using direct interest and deemed interest divided by total issued shared) 0: No independent directors. 1: One-third of the directors are independent directors. 2: More than one-third of the directors are independence. 0: No separate chairman and CEO. 1: Chairman and CEO are separate. 2: Chairman is independence. 0: Largest director s shareholding is more than 10% of issued capital. 1: Largest director s shareholding is 5%- 10% of issued capital. 12

17 Category Governance Standard Governance index 2: Largest director s shareholding is below 5% of issued capital. Audit Existence of audit committee 0: No audit committee. 1: Audit committee exists. 2: All members of audit committee are independence. Frequency of audit committee meeting 0: No audit committee. 1: Frequency of audit committee meeting is below and equal 4 times per year. 2: Frequency of audit committee meeting is more than 4 times per year. Expertise of audit committee 0: No member of audit committee has accounting background. 1: At least one of audit committee has accounting background. 2: More than one of audit committee has accounting or business or law background. Audit report and auditor 0: Qualified opinion of audit report. 1: Unqualified opinion of audit report. 2: Auditor in engagement is big 4 audit firms (PwC, KPMG, E&Y, Deloitte) and unqualified opinion. Compensation Existence of compensation committee 0: No compensation committee 1: Compensation committee exists. 2: At least a half of members (50%) of compensation committee are independence. Director nominations Existence of nomination committee 0: No nomination committee 1: No nomination committee exists. 2: At least a half of members (50%) of nomination committee are independence. 13

18 c. Measure of ownership structure Even if different date occurs different ownership shareholding; however, the main ownership structure in Thailand appears to be stable. In this research, I use shareholder listing as of presented in 56-1 and SETSMART at the close-book date of shareholder meeting during I assume ownership structure to foreign ownership, which tis international investors, including foreign individual and foreign juristic person. My sample of tax aggressiveness, corporate governance and ownership structure cover data during from listed companies in Thailand. In order to study the effect of these, I apply panel regression with year dummy to deal with the data and also solve with heteroskedasticity problem or fixed effect problem (if any). 14

19 IV. HYPOTHESIS AND CONTROL VARIABLES The hypothesis and control variables are constructed as details below: Hypothesis I: High corporate governance index firms present a higher level of tax aggressiveness relative to their lower index counterparts ETR i,t = β 0 + β 1 CGI i,t + β k Control Variables i,t + Year Dummies + Ɛ i,t (1) I examine whether high corporate governance firm avoid more or less tax than lower ones using year dummies as presented in the regression model (1) above. For this model, where I indexes firm, t indexes year, the dependant variable ETR is effective tax rate, measurement of tax aggressiveness. ETR combines with GAAP ETR or GAAP ETR, and CGI is corporate governance index. Control variables are stated as following. I expect a negative sign of coefficient on CGI, which results from prior research, Minnick and Noga (2010), who find that good corporate governance will decrease tax expense. Another supporting from Wang (2010), she states that transparency firms avoid more tax according to less severe of agency conflict. The decrease of tax expense means higher level of tax aggressiveness. Hypothesis II: Higher ownership concentration affects a higher level of tax aggressiveness ETR i,t = β 0 + β 1 CGI i, t + β 2 FOREIGN i, t + β k Control Variables i,t + Year Dummies + Ɛ i,t (2) I examine the relation of ownership structure over foreign holding to tax aggressiveness using year dummies as presented in the regression model (2) above. Variable of ownership structure comprises with foreign structure (FOREIGN). I expect a positive sign of coefficient on foreign structure according to literature, research from Europe. Huizinga and Nicodeme (2003) present that the higher foreign ownership causes higher tax expense. 15

20 For control variables, it s includes in the model consisted with 2 main groups: firms tax planning opportunities and general variables. According to the first group of control variables, firms tax planning opportunities, I control the variables that are related to tax shield. The reason for including these variables in the regression is that tax shield is used for tax planning, limited by a firm s opportunities to gain profit from decreasing taxable income. Referred to literature (Minnick and Noga, 2010), I include debt ratio (DE), long-term debt divided by book value of equity to be control variable. Note that the level of debt results to agency problem. Coefficient of this debt ratio is expected to negative. I include capital intensity (PPE) to control for a firm s tax planning opportunities. This capital intensity is used as proxy of capital expenditure ratio which I expect to positive coefficient according to literature (Wang, 2010). Level of capital intensity can reflect tax level for example depreciation method for tax purpose. For Thai firms, intangible assets presented in financial statements are minority comparing with balance of property, plant and equipment. Therefore, even if ETR in some researches include with the amount of intangible assets, I find that it is insignificant to my analysis. The second set of control variables, general variables; I include profitability (ROA) to control overall performance, firm size (ASSETS) to be proxy size of the companies and book-to-market (BM) to control growth of the companies. Moreover, I control firms net loss (EARN), which can skew my result, by setting indicator variable. If earnings are positive, the variable equals to one and zero otherwise. I expect positive coefficient on ROA, ASSETS, BM and EARN following from prior literature. Firm size and growth play a role in tax management, which higher size and growth cause higher tax (Dyreng, 2008). Also higher performance of profitability reflects higher tax expense (Minnick and Noga, 2010). 16

21 Figure 2: Expected sign of determinants Variables Expected sign to ETR CGI - FOREIGN + DOMES - ROA + EARN + PPE + DE - BM + ASSETS + For the valuation implication of tax avoidance, I set the hypothesis as following: Hypothesis III: Price discount (premium) of being tax aggressive decreases (increases) with the level of corporate governance Q i,t = β 0 + β 1 ETR i,t + β 2 CGI i,t +β 3 (ETR I,t * CGI i,t ) + β k Control Variables i,t + Year Dummies + Ɛ i,t (3) I examine the relation of tax aggressiveness and corporate governance resulting to shareholder wealth where I indexes firm, t indexes year, the dependant variable Qi,t, is firm value, measured by proxy of firm valuation, as in the regression model (3) above. ETR is the level of tax avoidance, CGI is corporate governance index, and ETR * CGI is the interaction of tax aggressiveness and corporate governance index. Control variables are the same as the regression model (1) and (2). I use to proxy firm value and control for other factors that have been affected to firm value, corporate governance and tax aggressiveness. The primary variable is the interaction term, which I expect a negative coefficient on this term according to prior literature of Wang (2010). 17

22 Table II: Variables Definition The table shows definition and measurement of each variable, and presents variable dataset which is used as sample in this paper. Variables Variables definition Variables measurement Variables dataset GAAP ETR GAAP effective tax rate Percentage of tax expense / Pre-tax income SETSMART, Datastream Cash ETR Cash effective tax rate Percentage of cash tax paid / (Pre-tax income Special item) SETSMART, Datastream CGI Corporate governance An index that ranks the relative for Annual report and 56-1 from index corporate governance SETSMART, firms website FOREIGN Foreign ownership Percentage of foreign ownership shares over total outstanding shares Major shareholders from SETSMART Q (Market value of equity + Book value of liabilities) / Book value of total assets SETSMART, Datastream ROA Return on assets (Pretax income Extraordinary items) / Total assets SETSMART, Datastream EARN Earning Indicator variable coded as 1 if earning 0, zero otherwise SETSMART, Datastream PPE Property, plant and equipment Percentage of property plan and equipment / Total assets SETSMART, Datastream ASSETS Book value of total The logarithm of total assets SETSMART, Datastream assets DE Debt ratio Long-term debt / Book value of equity SETSMART, Datastream BM Book-to- market Book value of common equity / Market value of common equity SETSMART, Datastream 18

23 Summary statistics Table III shows the descriptive statistics of the sample 1,876 firms for all variables in Panel A. Panel B presents dependent variable, tax aggressiveness, and corporate governance index break by industry. Tax variables report that the average of GAAP effective tax rate is % whereas cash effective tax rate is %. Both of them shows that the effective tax rate for the company in Thailand less than corporate tax rate, 30%, therefore, I can conclude that the firms have tax avoidance in order to manage tax and pay less than the regulated rate. For proxy of firm value, Tobin Q, the average is According to the sample, the main ownership structure of Thai s firm is domestic holding computing as %, the rest is foreign holding as %. Corporate governance index, which ranks from 0-18, the average is as 61.71% of total score. As discrepancy in details of each factors, the significant factor results from audit factor which the mean is based on total score 8 (75.59%). Board of directors is 3.4 based on total score 6 (56.67%). The average remaining of compensation and director nomination is and based on the same total score 2, which can be computed as 43.3% and 39.8% respectively. For control variables, the average of ROA is 7.118, property plant and equipment per total assets is %, logarithm of assets is 9.581, ratio of long-term borrowing and equity is 1.04 and book to market ratio is Descriptive statistics by industry presents that the highest GAAP effective tax rate is resource industry while the highest of cash effective tax rate is property & construction. The highest score of corporate governance index is also resource industry, which is as 69.36%. 19

24 Table III: Summary Statistics The table in panel A: Descriptive Statistics shows summary statistics of my variables as see in Table 2: Variables Definition.Panel B: Sample Description by Industry reports number of sample, which I observe data in SET since It also reports sample description of GAAP effective tax rate (GAAP ETR), cash effective tax rate (CASH ETR), corporate governance index (CGI) and (Q) break by industry. Panel A: Descriptive Statistics Tax Variables N Mean SD 25th Median 75th GAAP ETR 1, CASH ETR 1, Firm Value Q 1, Firm-specific Variables FOREIGN 1, ROA 1, EARN 1, PPE 1, ASSETS 1, DE 1, BM 1, Governance Variables CGI 1, Board of directors 1, Audit 1, Compensation 1, Director nomination 1,

25 Panel B: Sample Description by Industry Mean Industry N % of obs. GAAP ETR CASH ETR CGI Q Property & Construction % Services % Technology % Consumer Products % Industrials % Resources % Agro & Food Industry % Total 1, % Table IV shows Pearson correlation matrix for all variables. For dependent variable, GAAP effective tax rate and cash effective tax rate has significant correlated although the correlation is quite low (0.473). The correlation also reports that corporate governance index has negative correlation to effective tax rate. 21

26 Table IV: Correlation Matrix The table presents Pearson correlation matrix for both dependent variables used in this analysis; GAAP effective tax rate (GAAP ETR) and cash effective tax rate (CASH ETR) and the independent variables; corporate governance index (CGI), foreign ownership (FOREIGN), (Q), return on assets (ROA), earning (EARN), property, plant and equipment (PPE), book value of total assets (ASSETS), debt ratio (DE) and book to market ratio (BM). *, **, *** indicate statistical significance at 10%, 5% and 1% respectively. GAAP ETR CASH ETR CGI FOREIGN Q ROA EARN PPE ASSETS DE CASH ETR 0.473*** CGI FOREIGN ** Q *** ROA 0.078*** 0.073*** 0.103*** 0.082*** 0.257*** EARN 0.215*** 0.262*** 0.112*** 0.039* 0.065*** 0.644*** PPE * *** *** *** ASSETS *** 0.228*** 0.055** 0.180*** 0.163*** 0.075*** DE ** BM *** *** *** ** **

27 V. EMPIRICAL RESULT a. Corporate governance and tax aggressiveness To see whether corporate governance will affect to tax aggressiveness, I begin by examining the panel data with solving heteroskedastic problem. Note that there is no fixed effect in this regression, so I use the panel least square model. The regression result shows in the table V. The coefficient of corporate governance index (CGI) is negatively related with both effective tax rate (GAAP ETR and CASH ETR), significantly. This result follows the literature of Wang (2010), whom states that the transparent firms with good governance avoid more tax relative to others. According to the empirical result, Thai firms which have good corporate governance tend to pay tax less than the firms which are bad corporate governance. Governance is a monitoring tool in order to control managerial tax avoidance behavior. GAAP effective tax rate is higher effect than cash effective tax rate as reporting in the higher coefficient. For control variable, the coefficient of profitability (ROA) is negatively related with the effective tax rate. In addition, results show that when company size (ASSETS) and capital intensity ratio (PPE) increase, the effective tax will decrease. This empirical study is also different from my expectation, which expect the positive coefficient of profitability (ROA), size (ASSETS) and capital intensity (PPE). The main reason for this difference is that large companies bear a smaller tax burden compared to small firms. Size does not seem to significantly influence profitability before taxes, it is suggested that large companies are more successful in avoiding taxes, possibly through tax planning and fiscal engineering (Nicodome, 2002). Another supporting from the research of Rego (2003), he find that larger, more profitable and multinational companies avoid more tax than other firms as decreasing in effective tax due to more resources and more incentive for tax planning. No significant on debt ratio and book to market ratio, which suggest that it has no impact to tax level. 23

28 Table V: Panel Regression of Tax Aggressiveness on Corporate Governance Index Panel Least Square with solving heteroskedasticity reports the regression between dependent variables, tax aggressiveness (GAAP ETR and CASH ETR) and main independent variable, corporate governance index (CGI) with control variables (ROA, EARN, PPE, ASSETS, DE, BM). *, **, *** indicate statistical significance at 10%, 5% and 1% respectively. The regression model is: ETR i,t = β 0 + β 1 CGI i,t + β k Control Variables i,t + Year Dummies + Ɛ i,t GAAP ETR Intercept *** (4.61) CGI *** (-3.31) ROA *** (-5.11) EARN *** (27.51) PPE *** (-3.47) ASSETS *** (-6.07) DE (0.61) BM (-0.00) CASH ETR (1.09) * (-1.75) *** (-10.83) *** (24.32) *** (-5.25) *** (-3.71) (0.55) (-0.76) Year Dummies Included Included N 1,876 1,876 24

29 Table VI that decomposes the relationship between corporate governance factors and effective tax rate, presents board of directors and audit committee are significant factors affecting to the tax level. I find that the companies which have transparent and good governance in related with board of directors and audit tend to have tax management in order to save tax, showing as less effective tax rate. However, for the remaining factors, compensation and director nomination, the results show significant affect only to cash effective tax. The evidence suggests that the relation between tax aggressiveness and corporate governance is not concentrated only in one governance factor. The impact of this corporate governance seems to be general broadly to tax management. Table VI: Panel Regression of Tax Aggressiveness on Corporate Governance Index by Factors Panel Least Square with solving heteroskedasticity reports the regression between dependent variables, tax aggressiveness (GAAP ETR and CASH ETR) and main independent variable, corporate governance index (CGI) break by 4 factors; board of directors, audit, compensation and director nomination, with control variables (ROA, EARN, PPE, ASSETS, DE, BM). *, **, *** indicate statistical significance at 10%, 5% and 1% respectively. Panel A: GAAP ETR GAAP ETR GAAP ETR GAAP ETR Intercept *** *** *** *** (5.45) (4.71) (5.72) (5.75) Corporate Governance Index -Board of Directors *** (-3.83) -Audit *** - - (-5.17) -Compensation (1.44) -Director Nominations (1.27) Control Variables Included Included Included Included Year Dummies Included Included Included Included N 1,876 1,876 1,876 1,876 25

30 Panel B: CASH ETR CASH ETR CASH ETR CASH ETR Intercept (1.38) (1.52) (0.42) (0.17) Corporate Governance Index -Board of Directors * (-1.94) -Audit ** - - (1.98) -Compensation *** - (-3.86) -Director Nominations *** (-3.77) Control Variables Included Included Included Included Year Dummies Included Included Included Included N 1,876 1,876 1,876 1,876 b. Corporate governance and ownership structure According to ownership structure, table VII shows that foreign holding has negative coefficient, significant related with the effective tax. The result shows that when international investors increase shareholding in the Thai firm, the effective tax rate tend to decrease, meaning the increasing of tax aggressiveness. This empirical result is quite interesting due to not consistent with prior research from Huizinga and Nicodeme (2003) whom research the relationship between foreign ownership and tax level in Europe. The reason for this difference is that the availability of foreign tax credits to multinational firms for foreign-source corporate income taxes may provide multinational firms with a comparative advantage to operate in high-tax countries. This tax credit, one of the tax avoidance measurements, will reflect that a higher national tax would raise the effective taxation of domestic firms while leaving the worldwide taxation of multinationals unchanged. 26

31 Moreover, the opposite sign between foreign holding and tax level also be supported by the study from U.S. Grubert, Goodspeed, and Swenson (1993) find that foreign-controlled companies in the U.S. have lower tax than domestic companies based on tax-return data with many possible explanations for example tax management through transfer pricing or profit shifting. Another research from Kinney and Lawrence (2000) also support that foreign-controlled firms pay less tax than other U.S firms. According to tax regulation in Thailand, multinational firms or firms with higher foreign holding could get benefit from tax credit. When the Thai firms pay dividend profit to foreign investors who incorporate in Thailand, tax rate is exempt when complying with the regulation (Royal Decree of Revenue Code no.10, 2500, Section 5/18) or apply for 10% withholding tax rate when Thai firms pay dividend to foreign investors who not incorporate in Thailand or exempt tax when investing in the firms under BOI. Both of regulation also supports that the companies with more foreign investors or foreign holding will take tax benefit to get lower tax rate than average company tax. Therefore, these tax aggressiveness procedures should lead to higher percentage of foreign holding. Other control variables present as the same as the regression in table V. 27

32 Table VII: Panel Regression of Tax Aggressiveness on Corporate Governance Index and Ownership Structure Panel Least Square with solving heteroskedasticity reports the regression between dependent variables, tax aggressiveness (GAAP ETR and CASH ETR) and main independent variable, corporate governance index (CGI) and ownership structure (FOREIGN) with control variables (ROA, EARN, PPE, ASSETS, DE, BM). *, **, *** indicate statistical significance at 10%, 5% and 1% respectively. The regression model is: ETR i,t = β 0 + β 1 CGI i, t + β 2 FOREIGN i, t + β k Control Variables i,t + Year Dummies + Ɛ i,t GAAP ETR Intercept *** (3.77) CGI *** (-7.87) FOREIGN *** (-8.68) ROA *** (-5.01) EARN *** (27.33) PPE *** (-3.67) ASSETS *** (-4.15) DE (0.51) BM (-0.81) CASH ETR (1.03) ** (-2.35) *** (-3.23) *** (-10.75) *** (24.28) *** (-5.06) *** (-3.32) (0.53) (-0.88) Year Dummies Included Included N 1,876 1,876 28

33 c. Corporate governance and firm value To see whether corporate governance and tax aggressiveness will result to firm value, I use Tobin Q as a proxy to run panel regression. Table VIII shows the result of relationship of governance to firm value. Panel least square with solving heteroskedastic problem is employed. Moreover, after checking fixed effect problem, this model shows that it exists. As comparing between fixed effect model and random effect model by Hausman test, the random effect model is appropriate. Therefore, I also run random effect model with robustness test by clustering standard error in firms in order to solve fixed effect and heteroskedastic problem. For panel least square model, corporate governance has significant effect to firm value consistent with prior study as Cheung, Jiang, Limpaphayom, and Lu (2010) study that governance positively relates to market valuation. In addition, the results also support the paper of Connelly, Limpaphayom, and Nagarajan (2012), which say that corporate governance, on average, is positively associated with Tobin s q based on the sample in Thailand, year ended Good corporate governance leads to better performance, then higher valuation than the companies that have nontransparent. Investors tend to add up price premium to firms with good governance, then lead to value enhancing in the firm valuation. For the effective tax both GAAP effective tax, cash effective tax, and also the interaction between effective tax and governance, I find that there is no significant impact to firm value in Thailand even if the empirical reports the same sign of coefficient as the interaction is negatively correlated as studied of Wang (2010). Return on assets (ROA), profitability ratio, shows positively correlated to the firm valuation. Other coefficients on the control variables are generally in the prediction. Increasing in growth, increasing in firm size (ASSETS), and also increasing in market capitalization of the firms (BM) lead to increase in firm value as. 29

34 Table VIII: Panel Regression of on Corporate Governance Index Panel least square and random effect model, clustered by firms, report the regression between dependent variables, (Q) as proxy of firm value and independent variable; ROA, EARN, PPE, ASSETS, DE, BM, effective tax rate (GAAP ETR and CASH ETR) and interaction of tax aggressiveness and corporate governance index (ETR*CGI). *, **, *** indicate statistical significance at 10%, 5% and 1% respectively. The regression model is: Q i,t = β 0 + β 1 ETR i,t + β 2 CGI i,t +β 3 (ETR I,t * CGI i,t ) + β k Control Variables i,t + Ɛ i,t Panel Least Square Model Random Effect Model Intercept 1.021*** 1.047*** 1.691*** 1.691*** (13.91) (14.02) (2.50) (2.50) CGI 0.005*** 0.005*** (2.64) (2.72) (0.43) (0.44) GAAP ETR 3.06e -04 (1.33) e -04 (1.07) - CASH ETR e-04** (2.26) e -04 (1.20) GAAP ETR*CGI -1.17e -05 (-0.66) e -07 (-0.04) - CASH ETR*CGI e -05 (-1.44) e -06 (-0.37) ROA 0.020*** 0.020*** 0.017*** 0.017*** (22.54) (21.51) (3.31) (3.33) EARN *** *** *** *** (-13.21) (-13.25) (-3.07) (-3.13) PPE 5.10e -04 ** 5.67e-04** (2.16) (2.38) (1.58) (1.60) ASSETS 0.041*** 0.038*** (5.36) (4.88) (-0.57) (-0.56) DE -2.52e e e -04 *** -2.98e-04*** (-0.71) (-0.73) (-3.02) (-3.04) BM *** *** *** *** (-33.92) (-34.04) (-3.37) (-3.37) 30

35 Panel Least Square Model Random Effect Model Year Dummies Included Included Included Included R-Square % 21.10% N 1,876 1,876 1,876 1,876 Decomposing corporate governance factors, table IX reports that the main governance, significant impacting to firm value is audit committee. For other factors; board of directors, compensation and director nomination, it has no significant effect to the firm value. As random effect model, the result reports the same sign of coefficient of independent variables and control variables. However, the effect of corporate governance is not significant correlated to firm valuation. Overall result is consistent with the research of Desai and Dharmapala (2005), indicates that as average tax avoidance is no significant effect to the firm value. However, the firm with wellgoverned should lead to positive relationship to the firm value. The findings show that the simple view of corporate tax avoidance as a transfer of resources from the state to shareholders is incomplete, given the agency problems characterizing shareholder-manager relations. 31

36 Table IX: Panel Regression of on Corporate Governance Index by Factors Panel least square and random effect model, clustered by firms, report the regression between dependent variables, (Q) as proxy of firm value and independent variable; ROA, EARN, PPE, ASSETS, DE, BM, effective tax rate (GAAP ETR and CASH ETR) and interaction of tax aggressiveness and corporate governance index (ETR*CGI). This table separates corporate governance index by 4 factors; board of directors, audit, compensation and director nomination. *, **, *** indicate statistical significance at 10%, 5% and 1% respectively. Model 1: Panel Least Square Panel A-GAAP ETR Panel B-CASH ETR Intercept 1.010*** 1.135*** 1.014*** 1.028*** 1.034*** 1.144*** 1.035*** 1.048*** (13.74) (14.59) (13.66) (13.53) (13.87) (14.57) (13.79) (13.66) Corporate Governance Index -Board of Directors (0.84) (0.80) Audit *** (4.28) *** (4.12) - - -Compensation ** (2.43) *** (2.61) - -Director Nominations (0.78) (0.93) Control Variables Included Included Included Included Included Included Included Included Year Dummies Included Included Included Included Included Included Included Included N 1,876 1,876 1,876 1,876 1,876 1,876 1,876 1,876 32