CC BY
104
DOI https://doi.org/10.18551/rjoas.2018-11.23
PROFIT SHIFTING DETERMINANTS AND TAX HAVEN UTILIZATION: EVIDENCE FROM INDONESIA
Lambok Randy Suqih*, Jasman Jasman
Faculty of Economic and Business, IKPIA Perbanas, Jakarta, Indonesia *E-mail: r.s.lambok.sitompul@gmail.com ORCID: 0000-0002-9563-7682
ABSTRACT
Tax avoidance arrangement predominantly refers to multi-region schemes that fall under the grey area using a profit shifting scheme between high tax and negligible tax jurisdictions considered as tax havens. Multinational firms retain the incentive to enlarge their after-tax profits by implementing profit shifting from subsidiaries incorporated in high tax jurisdictions to subsidiaries in tax havens. The schemes to shift revenue including a higher proportion of debt to finance investments in high tax jurisdictions (thereby increasing deductible interest payments) and manipulation of transfer prices in intrafirm transactions Tax havens are multiple jurisdictions characterized by negligible or zero to low single digits income taxes. Moreover, tax havens deploy a degree confidentiality of financial and taxation information. The effects of tax haven utilization recently have start raising significant tax compliance and procedural fairness issues. This research provides empirical evidence of multinationality, transfer pricing aggressiveness, thin capitalization, intangible assets and size which represent the key indicators of profit shifting and tax haven utilization. Based on a sample of 85 listed public firms in Indonesia over the 2012-2016 period, we find that multinationality, transfer pricing aggressiveness, and intangible assets have positive effects on tax haven utilization and they are statistically significant. Meanwhile, our calculation results obtain thin capitalization and size have non-significant correlations with tax haven utilization.
KEY WORDS
Profit shifting, multinationality, transfer pricing, thin capitalization, tax havens.
Interdependence between countries followed by the rapid increase of trade and economic relations especially in the field of capital has led to a development of a new order in the global economy, namely the emergence of global economic unification with the trend toward regionalization and globalization. Globalization is described as an ongoing process of economic interdependence among the countries in the world with a feature of the rapid growth of financial transactions and international trade, especially among transnational firms, the wave of foreign direct investment which has broad support from transnational firms, the emergence of global markets, and the dissemination of technology and ideas as a result of the rapid expansion of transportation and communication systems across the world. (Singh, 2008).
Foreign Direct Investment (FDI) is a form of foreign investment which involves in improving the welfare of host country because of the benefits associated with new innovations, new technologies, managerial techniques, skills development, capital raising, employment creation, and industrial sector development at host country (Wadhwa, 2011). Ecletic Theory, one of the most extensive and comprehensive FDI theories developed by Dunning (1997), suggests that multinational firms impose FDI on host country is due to three motives, namely infrastructure procured information seeking, efficiency seeking proxied by exchange rates and macroeconomic stability as measured by inflation, and market seeking proxied by the size of the domestic market and the degree of economic openness. From several previous studies, FDI flows are broadly influenced by the behavior of multinational firms to gain comparative advantage.
The primary objective of a firm establishment is to increase shareholder wealth (Sharfman, 2013). Efforts to maximize the welfare of shareholders can be done by tax
planning that allows the taxpayer minimize the burden of tax payable through a scheme that is not contrary to the tax regulations (Darussalam, 2009). Tax planning schemes are mostly done by firms and the business world is a form of tax avoidance that is considered legal and is not considered to be contrary to moral in public view because it does not violate the provisions of law and legislation (Back, 2013). From this definition, it is described that tax evasion is a scheme in order to obtain benefits and benefits related to taxation by utilizing affiliated firms in tax haven country and weakness in the existing taxation system and regulation.
Along with the large amount of foreign investment coming into Indonesia, government revenues from taxes from multinational firms should be high. Aryanti (2016) on Liputan6.com noted Director General of Taxes had revealed that 2000 multinational firms operating in Indonesia do not pay the Income Tax Article 25 and Article 29 for 10 years for reasons of loss. The tax avoidance practices undertaken by these firms are carried out by transfer pricing scheme from Indonesia to another country.
The Director of Dissemination, Services and Public Relations of the Directorate General of Taxes stated that the 2000 firms indicated to be taxing because the reason for the loss was caused by 3 things. First, the firm is an affiliated firm whose parent firm is overseas so it is prone to occur transfer pricing process. There is a difference in tariff between Indonesia and partner countries so that they sell cheaply and buy raw materials at high prices. This difference in selling price and purchase price triggered the firm in Indonesia to suffer losses, but the overseas firms had profits. Second, thousands of multinational firms are losing money because many of them get tax incentive facilities such as tax holiday and tax allowance when applying for permission to the Investment Coordinating Board (Badan Koordinasi Penanaman Modal). Firms often increase the cost of purchasing capital goods. At a time when tax incentives have been exhausted the cost of purchasing high capital goods will result in high depreciation costs, so the burden and losses incurred by these firms are also higher. Third, the firm is often renamed in order to recover the tax incentives and re-recognize losses like the previous mode.
From economic and business perspective, the establishment of subsidiaries by multinational firms is to increase the competitive advantage of a firm. By using related parties, a multinational firm can expand the market in order to increase profits, find raw materials not obtainable in the country of origin, minimize production costs by finding cheap labor or minimize the cost of transporting raw materials and distribution costs of production in order to maximize the profit order can transfer losses incurred in the firm to other regions. Based on the facts described above, the problem that arises is a related party should work to enhance the competitive advantage of a firm while improving the welfare of the recipient country (Kogut, 1983), but in fact, a related party is often used to exercise tax avoidance mechanisms (Merks, et al., 2007).
UK Parliament (2013) denotes tax avoidance arrangement predominantly refer to multiregion schemes that fall under the grey area between acceptable tax avoidance and unacceptable tax avoidance using a profit shifting scheme between high tax and negligible tax jurisdictions considered as tax havens. Multinational firms retain the incentive to enlarge their after-tax profits by implementing profit shifting from subsidiaries incorporated in high tax jurisdictions to subsidiaries in tax havens. Schemes to shift revenue including a higher proportion of debt to finance investments in high tax jurisdictions (thereby increasing deductible interest payments) and manipulation of transfer prices in intrafirm transactions (Grubert & Mutti, 1991). The OECD, the G-20 and worldwide tax authorities have seriously scrutinized tax havens regarding to tax avoidance issue (Gravelle, 2013). Tax havens are multiple jurisdictions characterized by negligible or zero to low single digits income taxes. Moreover, tax havens also deploy a degree confidentiality of financial and taxation information through administrative practices or laws which prevent the exchange dan transparency of information (Wilson, 2009). Significantly, in terms of accounting earnings, debt location, investment allocation, and tax revenues tax havens play a vital role. Tax havens take place for particular arrangements of financing (e.g., assets protection and hedge funds) and businesses which considered to be driving force of the cost of capital and perform
as important offshore financial centres which facilitate the stream of capital between jurisdictions (Hanlon & Heitzman, 2010). In fact, many treasury operations in tax havens are exercised by multinational firms in order to accommodate the intragroup stream of capital without obstructions due to strict regulations and enforcement in accordance with capital management requirements and information stream (Dyreng & Lindsey, 2009; Gravelle, 2013).
Klassen & Laplante (2012) state multinational firms are technically capable to exercise profit shifting arrangements through the intentional intragroup transfer prices, preferential cost allocation, and tax motivated debt arrangements between variably taxed jurisdictions. Multinational firms have the opportunities of financial, regulatory and taxation arbitrage which are affected by courses of the shifting of debt and income in various tax jurisdictions. In particular, setting the prices of goods and services can obtain profit shifting which resulting in greater income to low-tax jurisdictions. They also mention that the preferential debt sourcing may create thinly capitalized structures in jurisdictions with high tax rates which resulting in greater tax deductions, such as loan fees and interest expenses, for the whole corporate as a group.
This research provides empirical evidence of multinationality, transfer pricing aggressiveness, thin capitalization, intangible assets and size which represent the key indicators of profit shifting by listed public firms in Indonesia. Also, this research is aim to encourage the Indonesian government to formulate regulations on the requirement for multinational firms in Indonesia to provide appropriate and reliable information on commercial nature of intragroup transactions in financial statements and tax documents. Last, this research could be potentially applied in other tax avoidance studies by developing the measurement of profit shifting determinants. This paper is consecutively organized into following order: introduction, literature reviews, research method, empirical results and conclusion.
LITERATURE REVIEW
The European Commission in its recommendation dated 6th December 2012, concerning aggressive tax planning, states that aggressive tax planning technically takes advantage of loopholes of tax system in a jurisdiction or between two or more tax systems for the intention of tax liability deduction, for example double nontaxation where firms abuse double-tax treaty networks which results in income not being taxed both in the state of residence and source or double deductions which result in loss deduction both in the state of residence and source. The OECD (2013) concludes in BEPS Report that some multinational firms have practiced tax avoidance and become more aggressive over time. The effects recently have start raising significant tax compliance and procedural fairness issues.
The existing studies provide evidence on how aggressively U.S multinational firms utilize profit shifting schemes through entities in variably taxed jurisdictions. The average global effective tax rates of U.S. multinational firms operating a subsidiary in a lax or nonexistent tax laws jurisdiction is around 1.5 % points lower than firms without tax haven incorporated operations and result in the reduction of corporate income tax by USD64 billion for the profitable corporations operating in tax havens during 1995-2007 period. The results underscore that U.S. multinational firms possess the incentives to sufficiently exercise tax motivated profit shifting and employ tax rate differentials between U.S. and the tax havens (Klassen & Laplante, 2012; Dyreng & Lindsey, 2009).
The existing research mentions that U.S. global firms possess opportunities to implement tax avoidance schemes. They found that the firms' effective income tax rate represents that aggressive strategies of profit shifting rely on tax havens and hybrid operations in variably taxed jurisdictions (Dyreng et al., 2008). Rego (2003) also reports that U.S. global firms with are more successful in practicing tax avoidance arrangements than firms operating only in U.S. In addition, Hanlon, Mills and Slemrod (2005) report that the levels of tax noncompliance by U.S. foreign controlled subsidiaries are more than double than domestic-controlled firms. Assuming that multinational firms implement effective tax
planning strategy among their entities, there is a tendency that firms with overseas subsidiaries are more likely to practice tax avoidance. Considering the research literatures, the hypothesis is proposed as follows:
H1: Firms with multinational characteristics are likely to utilize tax haven.
The term transfer pricing actually has a neutral concept. This term is used to distinguish intragroup transaction price from market price as if purchased from an unrelated third party (Wündisch, 2003). Basically, the effort to maximize profit is a natural thing considering that generally the main goal of a firm establishment is to generate profits as much as possible for shareholders. Viewing from the perspective of management accounting, the profit maximization of integrated multinational firms is achievable through aggressive and abusive transfer prices manipulation of products or services. This policy will increase profits in subsidiaries located in the jurisdictions with lower tax rate (Horngern, 2012). The existence of different tax rates in each country will lead to maximum after tax earnings of the corporations as a whole by declaring high returns in jurisdictions with low tax rates (Bernard & Weiner, 1990).
Tyrrall and Atkinson (1999) argued that globalization causes transfer pricing to be one of the most complex and significant taxation issues in the business world (Jarallah & Kanazaki, 2011). A survey conducted by Ernst & Young (2013) shows that tax-risk management is a top priority for multinational firms, caused by the increasing focus of tax authorities on transfer rates through checks and regulations. According to the survey, transfer prices have become one of the main causes of disputes between multinationals and tax authorities and Indonesia is on the list of the top five countries that impose sanctions on multinationals on transfer pricing.
According to Liebman and De Boeck (1988), prices imposed on multinational firms' transactions do not reflect a reasonable price and are often very different from the price to be charged if the transaction is made with a non-related party with the firm. Based on the description, the hypothesis is proposed as follows:
H2. Firms with aggressive transfer pricing characteristic are more likely to utilize tax haven.
Based on the theory of capital structure presented by Miller and Modigliani (1963), debt can be used to increase the value of the firm, because there are tax incentives received by the firm through the ability of interest expense to reduce taxable income. The condition that a firm uses more debt than capital as its funding source is called the practice of thin capitalization (OECD, 2012). The practice of thin capitalization can serve as one of tax avoidance arrangements (Lietz, 2014).
The definition of thin capitalization by OECD (2012) refers to a situation where firms are financed by higher debt levels compared to capital. This understanding is also in line with the definition of thin capitalization contained in the writings of Taylor and Richardson (2013) and Taylor, Tower and Van der Zahn (2010). Firms that apply this practice are also called highly-leveraged or highly-geared. The firm's strategy of obtaining capital has a noteworthy impact on the level of profit statements for tax purposes. The tax law allows for interest expenses already paid or which are still in the form of interest payable, as a deductible expense, when the measurement of fiscal profit is made. The higher the level of debt in the firm, the higher the interest expense to be paid. This results in lower fiscal profit (OECD, 2012).
Multinational firms often restructure their funding policies to maximize this profit. Not only can they construct a combination of debt and capital that results in tax efficiency in the source country, they can also influence the tax treatments imposed by the lender. For example, a firm may regulate the situation in which a lender may earn interest income in a tax jurisdiction that does not impose tax on interest income, or only impose a low tariff (OECD, 2012). Based on the description, the hypothesis is proposed as follows:
H3: Thinly capitalized firms are more likely to utilize tax haven.
The major problem regarding intragroup transfer of intangibles (i.e., patents, know-how, or scientific works) is that the intangible assets are very firm specific. This characteristic creates challenges for tax administrations to value the intangible assets at arm's length
principles and tax-planning opportunities for multinational firms by shifting profits to jurisdictions with lower tax rates. (Grubert and Mutti, 2007; Gravelle, 2010). U.S. Treasury Department (2007) asserts that the difficulties in determining the transactions involving intangible assets transfer and valuating arm's length prices and the variability in the interpretation of transfer pricing assessments result valuable intangible assets have a significant transfer risk in the form of transfer pricing aggressiveness. Intangible asset is a value-driver for multinational firms and highly mobile so it can easily be shifted (Grubert, 2003). Transfer pricing aggressiveness can be simultaneously obtained by the corporations in variably-taxed jurisdictions because the unavailability of mature markets and subjective valuation of intangibles (Dyreng et al., 2008). Considering the research literatures, the hypothesis is proposed as follows:
H4: Firms with greater investments in intangible assets are more likely to utilize tax haven.
Large size firms can rely on their tax planning schemes to achieve the competitive advantage over small size firms by using their resources to intentionally deduct their tax liabilities (Rego, 2003). Furthermore, Pittman & Fortin (2004) also mention larger firms with their capabilities can successfully reach economies of scale to obtain earnings and to expand or borrow capital at a lower cost than smaller firms. The firm size issues the capacity of profit shifting by utilizing tax havens. Richardson and Lanis (2013) with political cost theory show that firm size has a positive association with aggressive tax avoidance because large size firms have political power and superior economic relative over small size firms and they can accordingly distribute their tax liabilities. For instance, the case of manipulation by the Asian Agri Group was committed in 2002 to 2006 by 16 subsidiaries in Indonesia and 5 affiliated firms in tax havens is an evidence of aggressive tax avoidance behavior conducted by a large company. According to the investigation conducted by the Indonesian Directorate General of Taxes, the total loss to the country came to Rp 1.26 trillion (Dharmasaputra, 2014). Based on the description, the hypothesis is proposed as follows:
H5: Large size firms are more likely to utilize tax haven.
Cash flow from operations (CFO), net operating loss carryforwards (NOL), return on assets (ROA), the market-to-book ratio (MKTBK), industry sector (INDSEC) effects and year (YEAR) effects are incorporated as control variables.
We calculate CFO in the regression to indicate how a firm is likely to perform to meet its future obligations, including payroll, taxes and debts (Dechow, Kothari, & Watts, 1998). CFO possesses a positive effect on tax avoidance and it is significant (Kim, Li, & Zhang, 2011). Hanlon (2005) mentions that book-tax differences as disclosed in the financial statements provide the earnings persistence which is also determined by the components of cash flow contained in the current earnings. The earnings persistence has implications for the assessment of future earnings of a firm. We calculate cash flows from operations per sum of assets to measure CFO.
H6: Firms with greater CFO are more likely to utilize tax haven.
Firms can utilize net operating losses carryforward to the future year's profits as part of corporate tax planning activities to deduct tax liability. For multinational firms the losses can be transferred among group members located in various tax jurisdictions to be effectively used, this includes the earnings offset derived from financial arrangements (Dyreng et al., 2008). NOL is equal to 1 if a firm has a NOL balance in its financial statements, and 0 otherwise.
H7: Firms with greater NOL are more likely to utilize tax haven.
ROA is often used accounting measure of performance in financial research because it represents the ability of firms to generate profits from its total assets (Rowe & Morrow, 1999; Peng & Luo, 2000). Blouin and Larcker (2012) identify that ROA has a positive and significant association with tax avoidance. The availability of resources encourages profitable firms to establish and rely on entities primarily as offshore financial centers (Kim and Li, 2014). The operating performance and profitability in the regression are controlled by ROA. We calculate profit before tax per sum of assets to measure ROA.
H8: Firms with greater ROA are more likely to utilize tax haven.
MKTBK determines stock valuation and how the market views a firm's value. This variable is incorporated as a control for firm's growth in the regression model. The potential of incentives to establish tax haven entities describe that the operational and / or tax environments affecting high investment opportunities or growth firms are complex (Chen et al. 2010) and could affect the flow of capital and capital costs (Kim & Li, 2014). We calculate market value of equity per book value of equity to measure MKTBK.
H9: Firms with greater MKTBK are more likely to utilize tax haven.
INDSEC is equal to 1 in conformity with firm's type of business on Jakarta Stock Industrial Classification (Jasica) codes. This variable is included as a control of profit shifting fluctuation across various industry sectors in the regression model (Rego, 2003). In the regression model the variance in tax haven activities during the sample year 2012-2016 is controlled by YEAR.
METHODS OF RESEARCH
The sample consists of 85 listed public firms in Indonesia for the 2012-2016 period. We excluded financial firms, insurance firms, firms with no overseas subsidiaries, firms listed in the Indonesia Stock Exchange (IDX) after 2012 and firms not fully listed in 2012-2016 period. The financial and insurance firms were eliminated from the sample due to significant differences in the accounting standards and constraints. We collected the sample data from the financial statements publicly available on Indonesia Stock Exchange website.
In accordance with Richardson & Taylor's research (2015), the dependent variable is identified by THAV. We used three alternatives of tax haven utilization based on existing researches to obtain the robustness of the result. The first variable to indicate tax haven utilization is THAV1 in accordance with the research by Akamah, Hope, and Thomas (2014). THAV1 is equal to 1 if the firm has at least a subsidiary in the OECD (2006) tax haven list, and 0 otherwise. The second variable is THAV2 by reference to the study conducted by Desai et al. (2006). THAV2 is equal to 1 if a firm uses special purpose financing or insurance at least a subsidiary in the OECD (2006) tax haven list for treasury purposes, and 0 otherwise. The third variable is THAV3 by reference to the study by Desai & Dharmapala, (2009). THAV3 is a continuous variable and we calculate sum of subsidiaries in the OECD (2006) tax haven list per sum of subsidiaries to obtain the variable.
The independent variables consist of multinationality (MULTY), transfer pricing (TPRICE), thin capitalization (TCAP), intangible assets (INTANG) and firm size (SIZE). We calculate the sum of overseas subsidiaries per the sum of subsidiaries to obtain MULTY. TPRICE were hand-collected from the disclosures of related party transactions in financial statements in order to indicate the aggressiveness of transfer pricing. TPRICE is equal to 1 if the transactions with related parties are non-arm's length, and 0 otherwise. TCAP is equal to 1 if a firm's DER exceeds 1.5 and the net interest expense exceeds 50% of cash flow, and 0 otherwise. We calculate the sum of intangibles per the sum of assets to obtain INTANG. The firm size in this study refers to Hanlon (2005)'s research, where size is the natural logarithm of total assets.
Cash flow from operations (CFO), net operating loss carryforwards (NOL), return on assets (ROA), the market-to-book ratio (MKTBK), industry sector (INDSEC) effects and year (YEAR) effects are incorporated as control variables in the regression model.
The following formula describes the linear relationship:
THAVit = ait + p! MULTYit + p2 TPRICEit + p3 TCAPit + p4 INTANGit + p5 SIZEit + p6 CFOit + p7 NOLit +
p8 ROAit + p9 MKTBKit + p10-16 INDSECit + p17-22 YEARit + eit
Where: i = firms 1-85; t = financial years 2012-2016; THAV = tax haven indicator (THAV1-THAV3); MULTY = sum of overseas subsidiaries per sum of subsidiaries; TPRICE = a dummy variable, equal to 1 if the transactions with related parties are non-arm's length and 0 otherwise; TCAP = a dummy variable, equal to 1 if a firm's DER exceeds 1.5 and net interest expense exceeds 50% of cash flow, and 0 otherwise; INTANG = intangible assets per total
assets; SIZE = the natural logarithm of total assets; CFO = cash flow from operations per total assets; NOL = a dummy variable, equal to 1 if a firm has net operating loss carryforwards, and 0 otherwise; ROA = pre-tax income per total assets; MKTBK = market value of equity per book value of equity; INDSEC = a dummy variable, equal to 1 in conformity with the firm's type of business on Jakarta Stock Industrial Classification (Jasica) codes, and 0 otherwise; YEAR = a dummy variable, equal to 1 if the year falls within the specific year category, and 0 otherwise; and £ = the error term.
RESULTS AND DISCUSSION
Descriptive Statistics. Table 1 shows the descriptive statistics for the dependent variables (THAV1, THAV2 and THAV3), independent variables (MULTI, TPRICE, TCAP and INTANG, SIZE) and control variables (CFO, NOL, ROA and MKTBK). The dependent variable THAV1 has a mean of 0.746, showing that approximately 74,6% of the sample firms have at least one subsidiary incorporated in the OeCd (2006) tax haven list. Meanwhile, THAV2 has a mean of 0.129, indicating that around 12,9% of the firms in the sample use a special purpose financing or insurance subsidiary incorporated in a tax haven for treasury purposes. For the independent variables, MULTI, TPRICE, TCAP, INTANG and SIZE have means of 0.507, 0.795, 0.353, 0.032, and 28.862 respectively. The mean, median and standard deviation of SIZE show the range of the natural logarithm is various from minimum 20.531 to maximum 32.207 with standard deviation 1.749. The range of the control variables are also presented in Table 1.
Table 1 - Descriptive Statistics
Variable Mean Median Standard Deviation Minimum Maximum
THAV 1 0,746 1 0,436 0 1
THAV2 0,129 0 0,336 0 1
THAV 3 0,201 0,148 0,214 0 1
MULTY 0,507 0,455 0,320 0 1
TPRICE 0,795 1 0,404 0 1
TCAP 0,353 0 0,478 0 1
INTANG 0,032 0 0,107 0 0,736
SIZE 28,862 28,795 1,749 20,531 32,207
CFO 0,100 0,079 0,175 -0,291 1,630
NOL 0,148 0 0,356 0 1
ROA 0,069 0,044 0,164 -1,084 0,885
MKTBK 3,060 1,134 7,255 -2,939 62,931
Correlation Results. The Pearson correlation results in Table 2 indicates that THAV1 has significant correlations with MULTY and INTANG (p < 0.01 or better) and the control variables ROA and MKTBK (p < 0.01 or better). Meanwhile, THAV2 has significant correlations with the independent variables TCAP, INTANG (p < 0.01 or better) and SIZE (p < 0.05) and the control variables ROA and MKTBK (p < 0.05). THAV3 has significant correlations with the independent variables MULTY (p < 0.10 or better) and the control variables MKTBK (p < 0.05 or better). In fact, ROA and CFO have the highest correlation coefficient of 0.443 (p < 0.01). Finally, the variance inflation factors (VIFs) is calculated to estimate the regression models to test for signs of multi-collinearity between the explanatory variables. The overall correlation test results show that there are no correlations among independent variables that are higher than 0.90. This can be seen from the results of the correlation test shown in Table 2. Thus, it can be concluded that there is no multicollinearity among independent variables. In addition to the correlation matrix, this study examines the presence of multicollinearity by using Tolerance and Variance Inflation Factor (VIF) values. Figures commonly used to indicate the presence of multicollinearity are Tolerance <0.10 or VIF> 10. The results indicate none of the tolerances below 0.10 and VIF exceeded 10 for all explanatory variables. Therefore, there was no multicollinearity problem in this study.
RJOAS, 11(83), November 2018 Table 2 - Correlation Matrix
Variables THAV1 THAV2 THAV3 MULTY TPRICE TCAP INTANG SIZE CFO NOL ROA MKTBK
THAV 1 1
THAV 2 225*** 1
THAV 3 .551*** .086" 1
MULTY .156*** -.082* .392*** 1
TPRICE .106" .109** 0,013 0,042 1
TCAP -G:Q1Q .199*** -0:071 -.246*** 0,045 1
INTANG .132*" .118*** 0:Q69 -.088** .113** -0,004 1
SIZE 0:015 090** -0:Q4Q -0,033 .130"* -0,018 .156"* 1
CFO -G:Q42 -0,053 0:Q19 .222"* .102** -.128*** 0,047 0:Q16 1
NOL -0:015 -0:003 -0,023 -.102** 0,064 0,080 0,048 0,071 -0,041 1
ROA -.149*** -.090** 0:QQE .2S7"* 0,037 -.222*" -.119"* 0:Q4E .443"* ,130***
MKTBK -.143*** -.097** .083** .210*** .126"* .150*** -0,006 0,037 .371"* ,119***
Notes: ***Significant at 1%; **Significant at 5%; *Significant at 10%.
Regression Results. The result of Overall Fit of Model shows the value of Chi-Square is 106.078 with the degree of freedom (df) of 20 and the value of sig. of 0,000. In other words, there appear to be significant associations between independent variables (MULTY, TPRICE, TCAP, INTANG, SIZE), control variables (CFO, NOL, ROA, MKTBK, INDSEC1-6, and Year) and THAV1 or at least one of MULTY, TPRICE, TCAP, INTANG, SIZE and control variables that affects THAV1. Hosmer and Lemeshow's test shows the value of Sig. is 0.655. (> Sig .0.05) so it can be assumed the model can predict the observation value. In addition, overall percentage shows that the model accuracy rate is 77.6%.
The Variables in the Equation Table in table 3 presents the regression result with logistic regression model as follows:
Logitl" 71 ^ 1 = -19,737 + 2,552 MULTY + 0,718 TPRICE + 0,040 TCAP + 12,472 INTANG - 0,031 SIZE +
0,101 CFO-0,256 NOL-4,099 ROA-0,034 MKTBK - 17,856 INDSEC1 - 18,202 INDSEC2 - 18,582 INDSEC3 - 19,713 INDSEC4 - 18,752 INDSEC5 - 18,305 INDSEC7 - 18,981 INDSEC9 - 0,080 YEAR 2013 + 0,064 YEAR 2014 - 0,058 YEAR 2015 - 1,838 YEAR 2016
From the formula above, the following conclusions are drawn:
• The increase in the ratio of overseas subsidiaries to total number of subsidiaries by 1 point leads the increase of tax haven utilization of 2.552 times;
• The firm that disclosures the transactions with related party is non-arm's length has an odds ratio to utilize tax haven of 0.718 times;
• The increase in debt to equity ratio by 1 point leads to the increase of tax haven utilization of 0.040 times;
• The increase in the ratio of total intangible assets to total assets by 1 point leads to the increase of tax haven utilization of 12.472 times;
• The increase in the firm size by 1 point leads to utilize a tax haven incorporated firm of 0.031 times;
• The other variables remained constant.
The summary of the overall regression results is presented in Table 4. The summary presents the estimated value of the coefficient and the probability value of individual parameters from the regression test (Test Statistic t), as well as the significant values of the THAV1 - THAV2 from Logistic Regression Test. The coefficients for INDSEC and YEAR are not presented to be more concise. In order to test the research hypothesis, the dependent variable of THAV1 has been chosen as the main model. Meanwhile, the dependent variable THAV2 through THAV3 is the alternative measurements to test the robustness of the research main model.
RJOAS, 11(83), November 2018 Table 3 - Variables in the Equation
Variables in the Equation
G S.E. Wald df Slg. Exp(E) 95% C.l.for EXP(B)
Lower Upper
Step 1 ä MULTY 2,552 ,510 25,082 1 ,000 12,832 4,727 34,837
TPRICE ,718 ,328 4,803 1 ,028 2,051 1,079 3,899
TCAP ,040 ,309 ,017 1 ,898 1,041 ,568 1,907
INTANG 1 2,472 5,859 4,531 1 ,033 260955,348 2,680 2.530E + 1 0
SIZE -.031 ,085 ,1 36 1 ,71 2 ,969 ,821 1,144
CFO ,1 01 ,891 ,013 1 ,91 0 1,106 ,1 93 6,338
NOL -,25ö ,387 ,437 1 ,508 ,774 ,363 1,053
ROA -4,099 1,502 7,450 1 ,006 ,017 ,001 ,31 5
MKTBK -,034 ,024 2,009 1 ,156 ,966 ,922 1,01 3
INDSEC1 -1 7,850 17299,292 ,000 1 ,999 ,000 ,000
INDSEC2 -1 8,202 17299,292 ,000 1 ,999 ,000 ,000
INDSEC3 -1 8,582 17299,292 ,000 1 ,999 ,000 ,000
INDSEC4 -19,713 17299,292 ,000 1 ,999 ,000 ,000
INDSEC5 -1 8,752 17299,292 ,000 1 ,999 ,000 ,000
INDSEC7 -1 8,305 17299,292 ,000 1 ,999 ,000 ,000
INDSEC9 -1 8,981 17299,292 ,000 1 ,999 ,000 ,000
Th2013 -.080 ,422 ,036 1 ,850 ,923 ,404 2,110
Th2014 ,064 ,427 ,023 1 ,880 1,066 ,462 2,464
Th2015 -.058 ,434 ,018 1 ,894 ,944 ,403 2,208
Th2016 -1,838 ,402 20,878 1 ,000 ,1 59 ,072 ,350
Constant 1 9,737 17299,292 ,000 1 ,999 372924359,6
a. Variables) entered on step 1: MULTY, TPRICE, TCAP, INTANG, SIZE, CFO, NOL, ROA, MKTBK, INDSEC1, INDSEC2, INDSEC3, INDSEC4. INDSEC5, INDSEC7, INDSEC9, Th201 3, Th201 4, Th201 5. Th201 6.
Table 4 - Empirical Results
Variables Predicted Sign THAV 1 THAV 2 THAV 3
MULTY + 2,552*** -0,522 0,562498***
TPRICE + 0,718** 0,987* -0,305711**
TCAP + 0,040 0,978*** 0,129225**
INTANG + 12,472** 2,778* -0,176825
SIZE + -0,031 0,121 0,051167
CFO + 0.101 0,731 0,348001
NOL + -0,256 -0,643 -0,042507
ROA + -4,099*** -0,020 -0,232763**
MKTBK ? -0,034 -0,279** -0,004150
INDSEC ? Yes yes yes
YEAR ? Yes yes yes
Notes: ***Significant at 1%; **Significant at 5%; *Significant at 10%.
MULTY: Multinationality; TPRICE: Transfer Pricing Aggressiveness; TCAP: Thin Capitalization; INTANG: Intangible Assets; SIZE: Firm size; CFO: Cash flow from operating activity; NOL: net operating loss carry forwards; ROA: Return on Assets; MKTBK: market-to-book ratio; INDSEC: industry sector effects; YEAR: year effects.
H1: Firms with multinational characteristics are likely to utilize tax haven.
The influence of MULTY on THAV1 is indicated by the coefficient 2.552, means that MULTY has a positive effect on THAV1. The calculation result obtained a significance value 0.000 (p < 0.01), it means the positive effect is statistically significant. Thus, it can be concluded that H1 is supported. Indonesian multinational firms may have an opportunity to gain a greater benefit from the value of a deductible interest expense and take advantage of different tax rates between taxation jurisdictions. Indonesian firms also may benefit when receiving loans from subsidiaries located in jurisdictions that impose a lower corporate tax rate. This result is in line with research results conducted by Taylor & Richardson (2013) and Hanlon, Mills, & Slemrod (2007) which mention that firms conducting large scale of multinational operations are more likely to utilize a tax haven incorporated firm.
H2: Firms with aggressive transfer pricing characteristic are more likely to utilize tax haven.
The regression correlation for TPRICE shows a positive sign indicated by the coefficient 0.718 and a significance value 0.028 (p < 0.05). It means TPRICE has a positive
effect on THAV1 and it is statistically significant. Thus, H2 is supported. Transfer pricing could be market based or alternatively nonmarket based. In the case of nonmarket based pricing, goods or services are either transferred above or below the prevailing market prices. It is predicated on the aggressive tax avoidance scheme. This is the instance when the concern for fairness and transparency arises. Transfer pricing schemes are designed to determine the amount of profit or loss that is attributable to the intragroup economic activities. Regrettably, these arrangements for all intents and purposes are artificial without economic substance, principally aimed to avoid payment of taxes. Such aggressive tax activities have been strongly criticised for causing the tax base erosion which tends to prevent the host country to mobilize domestic resources for development (Sundaram, 2012). It is reported by the Chinese government that about 60% of company losses recorded by the country were false, thereby triggering a tax loss of over 30 billion Yuen (US$4.39 billion) per annum (Chang & Jin, 2016).
H3: Thinly capitalized firms are more likely to utilize tax haven.
The influence of TCAP on THAV1 is indicated by the coefficient 0.040, means that TCAP has a positive effect on THAV1. But, the calculation result obtained a significance value 0.898 (p < 0.05), it means the positive effect is statistically non-significant. Thus, H3 is rejected. Based on the theory of capital structure presented by Modigliani & Miller (1963) debt can be used to increase the value of the company, because there are tax incentives received by the company through the ability of interest expense to reduce taxable income. In this case, thinly capitalized firms have an opportunity to exercise tax avoidance schemes (Lietz, 2013). Ogundaja & Onakoya (2016) find that firms in developing countries are favourably financed by equity and do not use excessive debt. The regression result reflects the behaviour of firms in developing countries regarding capitalization. Besides, the Minister of Finance of Indonesia in October 1984 issued a decision Number 1002/KMK.04/1984 regulating the determination of deduction of interest expenses in income tax calculation which limited a debt-to-equity ratio (DER) of 3:1. Considering the investment climate in Indonesia, the implementation of the regulation was postponed through decision Number 254/KMK.04/1985 dated 8 March 1985. Thus, there was no provision of limitation on thin capitalization in Indonesia until 2016. The firms freely determined the proportion of debt to equity ratio. Finally, the Minister of Finance officially issued a regulation number PMK-169/PMK.010/2015 dated 9 September 2015 regulating the determination of comparative amount between firms' debt and equity in income tax calculation, which is effective since the fiscal year 2016 with debt to equity ratio level set as high as 4: 1.
H4: Firms with greater investments in intangible assets are more likely to utilize tax haven.
The influence of INTANG on THAV1 is indicated by the coefficient 12.472, means that INTANG has a positive effect on THAV1. The calculation result obtained a significance value 0.033 (p < 0.05), it means the positive effect is statistically significant. Therefore, H4 is supported. The existing studies conducted by Conover and Nichols (2000) and Treasury (2007) saying that the preferential locations of intangible assets in jurisdictions with high tax rate decrease the corporation income tax liability, and the increase in flow of intangible assets in the financial statements can be the indicator. This event may lead to the manipulation of transfer pricing. A tax haven incorporated subsidiary can acquire the rights of a valuable intangible property at a favourable price and engage with the operational subsidiaries in high-tax jurisdictions in a cost sharing agreement with a hybrid structure to facilitate the payment of royalties' benefits that are attributable to the assets. Thus, the deductible expense paid to tax haven will be eligible to zero or low tax rate (Mutti & Grubert, 2007).
For instance, in a 2012 U.S. Senate enquiry outlined the abuse of transfer-pricing involving Microsoft. Microsoft exercised a tax avoidance scheme by transferring significant rights of intellectual property to Ireland and Singapore which resulting payments amounted to $US2.8 billion and $US1.2 billion respectively. However, both subsidiaries still generated revenues of $US9 billion and $US3 billion from the transactions in a single year. In addition, Microsoft undertakes over 85% of research and development in U.S. (Levin 2012).
H5: Large size firms are more likely to utilize tax haven.
The regression correlation for SIZE shows a negative sign indicated by the coefficient -
0.031.and a significance value 0.969 (p < 0.05). It means SIZE has a non-significant negative effect on THAV1. This result indicates that H5 is rejected. The empirical result illustrates that firm size does not affect tax haven utilization in Indonesia. Considering that aggressive tax avoidance behavior in Indonesia is generally carried out either passively or actively, for small size firms can either go through accounting fraud, tax management or by not paying and filing taxes which they are liable. While aggressive tax avoidance conducted by large firms tend to use a strategy that is still in the gray area, such as the practices of double deduction, double non-taxation, transfer pricing and treaty shopping. Researches conducted by Kim and Limpaphayom, (1998), Derashid et al. (2003), Rego (2003), and Hanlon (2005) show that firm size is not associated with aggressive tax avoidance.
CONCLUSION
This research is purposed to study and provide empirical evidence of multinationality, transfer pricing aggressiveness, thin capitalization, intangible assets and size which represent the indicators of profit shifting towards tax haven utilization by listed public firms in Indonesia. We find that multinationality, transfer pricing aggressiveness, and intangible assets have positive effects on tax haven utilization and they are statistically significant. Meanwhile, our calculation results obtain thin capitalization and size have non-significant correlations with tax haven utilization.
Further research is expected to use more extensive and significant proxies in studying tax haven utilization towards audit, intangible assets, transfer pricing aggressiveness and thin capitalization. Second, further research is expected to use data owned by the Indonesian Directorate General of Taxes. Third, future research is expected to study the determinants of tax haven utilization in other countries and contributes to improve anti-tax avoidance regulations which provide the guidelines on the determination of arm's length prices, documentations and other practical issues relevant to the transfer pricing transactions.
REFERENCES
1. Akamah, H., Hope, O.-K., & Thomas, W. B. (2015). Tax Havens and Disclosure Aggregation. Working Paper.
2. Armstrong, C. S., Blouin, J. L., Jagolinzer, A. D., & Larcker, D. F. (2015). Corporate Governance, Incentives, and Tax Avoidance. Journal of Accounting and Economics, 60(1), 1-17. http://doi.org/10.1016/jjacceco.2015.02.003
3. Ariyanti, Fiki. (2016). "2000 Perusahaan Asing Gelapkan Pajak Selama 10 Tahun". Liputan 6. http://bisnis.liputan6.com/read/2469089/2000-perusahaan-asing-gelapkan-pajak-selama-10-tahun
4. Back, Foster, P. (2013). "Avoiding Tax May be Legal, but Can It Ever Be Ethical?". The Guardian. ISSN 0261-3077.
5. Baltagi, Badi H. (2005). Econometric Analysis of Panel Data Third Edition. England:John Wiley & Sons Ltd, The Atrium, Southern Gate
6. Bevan, A. A., & Estrin, S. (2004). The Determinants of Foreign Direct Investment Into European Transition Economies. Journal of Comparative Economics, 32(4), 775-787. http://doi.org/10.1016/jjce.2004.08.006
7. Blouin, J., Huizinga, H., Laeven, L., & Nicodéme, G. (2014). Thin Capitalization Rules and Multinational Firm Capital Structure. Taxation Papers, 42, 1-47. doi:10.2778./32784
8. Chen, S., X. Chen, Q. Cheng, and T. Shevlin. (2010). Are family firms more tax aggressive than nonfamily firms?. Journal of Financial Economics, 95 (1): 4161.
9. Coleman, N., & Bean, E. J. (2006). Tax Haven Abuses: the Enablers, the Tools and Secrecy Minority & Majority Staff Report Permanent Subcommittee. Office.
10. Desai, M., & Dharmapala, D. (2006). Earnings management and corporate tax shelters. Harvard University. http://doi.org/http://dx.doi.org/10.2139/ssrn.884812
11. Dharmapala, D., & Hines, J. R. (2009). Which countries become tax havens? Journal of Public Economics, 93(9-10), 1058-1068. http://doi.org/10.1016/jjpubeco.2009.07.005
12. Dharmasaputra, M. (2014). Key Witness: An Investigation of Indonesia's Largest Ever Tax Scandal. Indonesia: PT Tempo Inti Merdeka Harian, Jakarta
13. Dunning, J. H., (1977). Trade, location of economic activity and the MNE: A search for an eclectic approach in Ohlin, B., Hesselborn, P. O. and Wijkman, P. J. (eds.) The international allocation of economic activity. England: Macmillan, London.
14. Dyreng, S. D., Hanlon, M., & Maydew, E. L. (2008). Long-Run Corporate Tax Avoidance. The Accounting Review, 83(1), 61-82. http://doi.org/10.2308/accr.2008.83.1.61
15. Dyreng, S. D., & Lindsey, B. P. (2009). Using Financial Accounting Data to Examine the Effect of Foreign Operations Located in Tax Havens and Other Countries on U.S. Multinational Firms' Tax rates. Journal of Accounting Research, 47(5), 1283-1316. http://doi.org/10.1111/j.1475-679X.2009.00346.x
16. Erickson, M., Hanlon, M., & Maydew, E. L. (2004). How Much Will Firms Pay For Earnings That Do Not Exist? Evidence of Taxes Paid on Allegedly Fraudulent Earnings. Accounting Review. http://doi.org/10.2308/accr.2004.79.2.387
17. Frank, M. M., Lynch, L. J., & Rego, S. O. (2009). Tax Reporting Aggressiveness and Its Relation to Aggressive Financial Reporting. In Accounting Review (Vol. 84, pp. 467496). http://doi.org/10.2308/accr.2009.84.2.467
18. Gravelle, J. G. (2009). Tax Havens: International Tax Avoidance and Evasion. National Tax Journal, 62(4), 727-753. http://doi.org/10.2307/41790645
19. Gravelle, J. G. (2015). Tax Havens: International Tax Avoidance and Evasion. Congressional Research Service
20. Grubert, H., & Mutti, J. (1991). Taxes, Tariffs and Transfer Pricing in Multinational Corporate Decision Making. The Review of Economics and Statistics, 73(2), 285- 293. http://doi.org/10.2307/2109519
21. Gujarati, Damodar N. (2004). Basic Econometrics (4th edition). New York: McGraw-Hill Firms.
22. Harris, D., Morck, R., Slemrod, J., & Yeung, B. (1993). Income Shifting in U.S. Multinational Corporations. University of Chicago Press. http://doi.org/10.3386/w3924
23. Hodge, Margareth. (2013). MPs Publish report on Google's tax avoidance. Commite of Public Account. UK Parliament.
24. Klassen, K. J., & Laplante, S. K. (2012). The Effect of Foreign Reinvestment and Financial Reporting Incentives on Cross-Jurisdictional Income Shifting. Contemporary Accounting Research, 29(3), 928-955. http://doi.org/10.1111/j.1911-3846.2011.01136.x
25. Kogut, Bruce. (1983). Foreign Direct Investment as a Sequential Process, The Multinational Corporation in the 1980s. MIT Press.
26. Maftuchan, AH, Wiko, Saputro. (2013). Evaluasi Realisasi Penerimaan Pajak 2013: Berada Pada Titik Terendah Sejak 2011. Prakarsa.
27. Merks, Paulus. (2007). "Categorizing International Tax Planing" dalam Fundamental International Tax Planing, Raffaele Russo (ed). IBFD. Hal 66-69.
28. Needham, C. (2013). Library Briefing Corporate Tax Avoidance by Multinational Firms. Library of the European Parliament, 1-5.
29. OECD. (2012). Dealing Effectively with the Challenges of Transfer pricing, Paris: OECD Publishing.
30. OECD. (2013). Public Consultation: Draft Handbook On Transfer Pricing Risk Assessment. Paris: OECD Publishing
31. OECD. (2013). Public Consultation White Paper On Transfer pricing Documentation. Paris: OECD Publishing
32. OECD. (2006). The OECDs Project on Harmful Tax Practices: 2006 Update on Progress in Member Countries. Washington DC: OECD.
33. OECD. (2009). Countering Offshore Tax Evasion. Washington DC: OECD.
34. Omar, N., & Zolkaflil, S. (2015). Profit Shifting and Earnings Management through Tax Haven Subsidiaries: An Exploratory Analysis of Multinational Firms. Procedia Economics and Finance, 28(April), 53-58. http://doi.org/10.1016/S2212-5671(15)01081-3.
35. Rego, S. O. (2003). Tax-Avoidance Activities of U.S. Multinational Corporations. Contemporary Accounting Research, 20(4), 805-833. http://doi.org/10.1506/VANN-B7UB-GMFA-9E6W.
36. Richardson, G., & Taylor, G. (2015). Income Shifting Incentives and Tax Haven Utilization: Evidence from Multinational U.S. Firms. International Journal of Accounting, 50(4), 458-485. http://doi.org/10.1016/j.intacc.2015.10.001
37. Sharfman, B. S. (2013). Shareholder Wealth Maximization and Its Implementation Under Corporate Law. Florida Law Review, 66(1), 389-432.
38. Shevlin, T. (2007). The Future of Tax Research: From an Accounting Professor's Perspective. The Journal of the American Taxation Association, 29(2), 87-93. http://doi.org/10.2308/jata.2007.29.2.87
39. Singh, K. (2008). Taking It Private: The Global Consequences of Private Equity. Corner House Briefing Paper. 37, 1-26. http://www.thecornerhouse.org.uk/
