The impact of profit shifting regulation
on corporate profit margins:
a global investigation
University of Groningen
June 19th, 2015
Master of Science:
International Financial Management
Written by:Lars Polke
S 2164647
lars.polke@googlemail.com
Supervisor:
Dr. H. Vrolijk
Abstract:This master thesis studies the impact of profit shifting regulation on the profitability of
multinational corporations. I investigate the effectiveness of the two dominant regulations, thin-capitalization restrictions and transfer pricing regulation.
The obtained evidence from the sample of 28,966 international subsidiaries from 38 countries, indicates that thin-capitalization restrictions reduce corporate profit margins, and that this reduction is more severe in countries characterized by high tax rates. Further, the regulation of transfer pricing regulation also impacted corporate profitability negatively, but to a minor extent. Again, this reduction in profit margins is larger in countries with high tax rates.
Table of contents
1. Introduction 1
2. Literature 5
2.1 Internal debt financing 5
2.2 Transfer pricing 6
2.3 Related literature and findings 7
3. Hypotheses development 10
3.1 Thin-capitalization rules and profit margins 10 3.2 Transfer pricing regulation and profit margins 11
3.3 Tax rate sensitivity 12
4. Methodology 14
4.1 Dependent variable 15
4.2 Independent variables 15
4.2.1 Thin-capitalization rules 15
4.2.2 Transfer pricing regulation 16
1
1. Introduction
The onset of globalization has led rise to multinational corporations which are firms that operate in their domestic home-country, but additionally have operations outside national borders. Each of these foreign subsidiary of the firm is taxed in the country of residence which implies that the collective corporate taxes burden is the sum of all taxes being paid to multiple national
governments.
Since the tax rate is solely national concern, different countries apply different tax rates according to their domestic preferences and demands. This implies that subsidiaries of a multinational corporation are taxed at different rates which creates an incentive for firms to shift profits into low tax jurisdictions in order to minimize the global tax burden, and thus reduce corporate expenses.
Not surprisingly, the ability of profit shifting creates an advantage for multinational corporations towards their domestic rivals which are unable to employ these tax- reducing techniques
(Bucobetsky and Haufler, 2008).
2
investment. The authors obtain evidence that increased tax rates abroad result in reduced corporate investments in this country. Demirguec- Kunt and Huizinga (2001) investigate the profitability of banks across several countries influenced by the applicable domestic tax rate, and find that firms which are exposed to higher tax rates are less profitable. Lastly, Bartelsman & Beetsma (2003) concluded in their study that multinational corporation tend to report less added value in countries which inhabit higher tax rates.
Given these findings in the literature, there is solely limited evidence which applies a global approach. In addition, the underlying issue is also relevant in a professional context, since these finding have implications for multinational corporations and their respective managers.
Multinational firms are likely to suffer from reduced competitive advantages relative to their international rivals if we regard the global competition. This lower performance might lower the development of R&D, reduced corporate investment in the restricting country, and also lead to disinvestments which imply lower national tax revenues for governments.
Based on this theoretical and practical background, this study intends to provide updated
evidence on the effectiveness of profit shifting regulations, and expand the empirical findings to a global context. Further, this study investigates the resulting impacts of these regulations on firms’ profitability of multinational corporations which aims to provide new statistical insights for professionals, regulators, and academics alike.
To understand how governments regulate profit shifting of multinational corporations, I investigate the two main restrictions, thin-capitalization rules and transfer pricing regulations. Thin-capitalization rules are restrictions which are employed to combat internal debt financing between two subsidiaries, or the subsidiary and the parent. Internal debt financing is a method of profit shifting: a subsidiary in a low-tax jurisdiction extends debt to another subsidiary which is exposed to a higher tax rate. Now, the higher taxed subsidiary repays the debt, and reduces its tax burden since these debt payments are tax-deductible. Effectively, the subsidiary reduces its own profits through these debt expenses, but increases the overall profit of the firms since the
repayment of the debt is taxed at a lower rate in the country of the other subsidiary (Desai, Foley, and Hines, 2004).
3
haven ratio, is not allowed for tax deduction. Since the firms are not able to expense these excessive debt payments, the previous tax advantage is eliminated, and the government effectively limited profit shifting of multinational corporations.
The second governmental regulation aims to counteract the arbitrarily assigned transfer prices of intra-firm trade which is used as a channel to shift profits into lower tax jurisdictions. More precisely, subsidiaries in high-tax jurisdictions export their goods and services at low prices with small profit margins to other affiliated subsidiaries which results overall in small reported profits. Sequentially, the lower profits imply that this subsidiary effectively reduces its tax burden in this given country. The remaining actual share of profits is then realized by subsidiaries in low-tax jurisdictions which consequentially are taxed at a lower tax rate (Clausing, 2003). Ultimately, the multinational corporation assigns prices in such a way that profits are accounted for in low tax jurisdictions, and therefore achieve a reductions of the overall tax burden for the collective corporation.
In order to circumvent these method of profit shifting, the government establishes an allowable range of prices for the exchange of intra-firm goods and services. These prices are labeled arm-lengths prices, and are established through reference to similar and comparable transaction between unrelated parties. Ultimately, these restriction hinder firms to assign arbitrary prices to intra-firm trade which limits their ability to shift profits, and thus to avoid domestic taxation.
Beyond the impact on corporate profit margins, the effectiveness of the implemented regulation itself is not the only determining factor of profitability, but also the national statutory tax rate will moderate the intensity resulting from the restrictions. Since these regulations restrict profit
shifting and force multinational corporations to realize a certain amount of profits domestically, the present tax rate in this country also determines the extent of the corporate tax burden. Higher taxes would result in larger costs for multinational corporations, and effectively reduce their respective profit margins. Hence, not only the increase of regulations will theoretically reduce profit margins, but also will this effect be intensified by higher statutory tax rates.
4
28,966 multinational corporations from 38 different countries, and investigate each subsidiary’s profit margin in order to determine a reduction in profitability for the firm.
The multiple regression analysis provided evidence for the effectiveness of thin- capitalization regulation, and demonstrated that that increasing thin- capitalization restrictions reduce the profitability of multinational corporations. Further, this negative impact is more severe if the country simultaneously employs a high corporate tax rate.
In addition to this evidence, the effectiveness of transfer pricing regulation was also found to be statistically significant which implies that increased transfer pricing regulation impacts corporate profitability, but to a minor extent. Further, empirical evidence revealed that this impact is has a larger negative influence for countries with high national tax rates.
5
2. Literature
As outlined in the introduction, international differences in taxation have led to the rise of profit shifting by multinational corporations in order to reduce their effective tax burden overall. In this literature part, I will outline the theoretical means by which corporations achieve profit shifting, highlight the respective economic implications, and present the governmental regulation which aims to circumvent these activities. Additionally, I will provide several relevant findings from the existing literature in order the place the current study in perspective. Moreover, I will derive testable hypotheses based on theoretical argumentation and the previous findings in the literature. The two dominating means by which multinational corporations achieve profit shifting are internal debt financing, and transfer pricing.
2.1 Internal debt financing
Internal debt financing in the broadest sense implies that a multinational corporation adopts certain global capital structure which redistributes realized profits to different countries, thus altering the tax burden in the respective countries. The aim of redistributing the profits is the reduction of the collective amount of taxes paid by a multinational corporation on a global level. More precisely, the multinational company creates internal debt between a subsidiary and the parent, or between two different subsidiaries. Given that debt expenses are commonly tax-deductible, the subsidiary in a high-tax jurisdiction can take on debt from a subsidiary in a low-tax country. Further, the subsidiary in the high-low-tax country is able to expense these debt
6
This opportunity to reduce the global tax burden is recognized, and respective means of transferring profits are implemented by multinational corporations. Consequentially, countries have tried to restrict this practice by multinational corporations since it erodes their respective tax revenues (Buettner et al., 2014). More precisely, governments have responded by implementing thin-capitalization rules. In the broadest sense, these rules aim to limit the interest deduction of inter-company debt provided by subsidiaries located in low-tax jurisdictions. In more detail, thin-capitalization rules establish a fixed debt-to-equity ratio which restricts the amount of debt of a subsidiary in relation to its equity. This ratio is commonly referred to as the safe haven or safe harbor ratio in the literature (OECD, 1987), and it establishes the proportion of debt relative to equity which is allowed for tax reduction. Sequentially, subsidiaries inhibiting a debt level below this threshold are allowed to use their interest expenses as means to reduce corporate taxes. Contrary, subsidiaries with excess amounts of debt which exceeds the legally fixed ratio are solely able to expense debt payments until the threshold is met. Any further debt expenses which exceed this threshold are not tax-deductible which limits the corporate ability to shift their international profits.
2.2 Transfer pricing
Transfer pricing is the second method of international profit shifting which I will discuss in this thesis. In the broadest sense, transfer pricing is the assignment of prices for intra-company trade. This technique of profit shifting centers on the global accounting practices which force firms to assign prices to goods and services when they are exchanged between a parent and its subsidiary, or between two affiliated subsidiaries. In particular, a subsidiary in a low-tax jurisdiction would charge higher prices for exports to other subsidiaries within the multinational corporation.
Contrary, subsidiaries located in high-tax jurisdictions would pursue the strategy of charging low prices for intra-firm exports to related subsidiaries. Either way, the multinational corporation achieves a shift in profits which will be implemented in such a way that subsidiaries inhibiting relatively low corporate tax rates to increase their accounting profits by assigning higher prices to intra-firm exports (Clausing, 2003). Sequentially, this implies that subsidiaries in low-tax
7
It is important to notice that the assignment of intra-firm profits does not necessarily reflect real pricing of each subsidiary but is rather applied arbitrarily. From a theoretical point of view, firms would actually achieve the lowest collective tax burden if intra-firm trade is priced in such a manner that all profits are realized in the subsidiary with the lowest tax rate, while all other affiliated subsidiaries simply break even in accounting terms.
Regarding this method of profit shifting from a nation’s point of view, it can be inferred that governments, especially from high tax jurisdictions, are opposed to these techniques because it erodes their tax revenues. Since nations fear declining financial means, governments throughout the world have implemented transfer pricing regulation in order to restrict the profit shifting of multinational corporations, and to maintain their corporate tax revenues. The majority of
countries around the globe have implemented so called arm’s length principles. These principles aim to restrict the arbitrary assignment of intra-firm prices for goods and services, and effectively establish a range of acceptable prices. Acceptable prices for the internal trade are created by comparable transactions between unrelated and independent parties. More specific, intra-firm prices are compared to prices of other firms from the same industry or field of business. If prices are similar and comparable with alike transactions between unrelated parties, these prices are allowed by the government, and they do not require further adjustment. Even though this appears simplistic, finding and comparing similar transactions of other corporations is rather complex since this information firm-specific and commonly kept secretly.
Consequently, the OECD has aimed to facilitate comparability, and thus established guidelines for transfer pricing. These guidelines determine documentation requirements for corporations which increase practical comparison, and are implemented by the majority of countries around the globe (OECD, 2010).
2.3 Related literature and findings
8
which center on the resulting impacts on corporate behavior which are caused by this regulation, and which impact multinational corporations in different ways.
Previous scientific literature presents evidence for the effectiveness of thin-capitalization rules (Buettner et al., 2012; Overesch and Wamser, 2007).
Buettner et al. (2012) investigate thin-capitalization rules and the respective impact on firms’ debt financing based on a global sample. Their study derives the conclusion that increased thin-capitalization rules reduce the corporate use of internal debt financing, and that firms instead make use of unrestricted debt from external providers. This leads to the inference that thin-capitalization rules are effective, since they trigger structural adjustments of multinational corporations.
Overesch and Wamser (2007) also investigates the effectiveness of thin-capitalization rules by investigating the capital structure of German multinational corporation, and by identifying respective changes after the government increases the strictness of thin-capitalization regulation . Their results indicate that governmental regulation effectively reduces the use of internal
borrowing, and therefore restrict profit shifting by multinational corporations. These studies hint that thin-capitalization regulations achieve a restriction of profit shifting in the investigated countries which is likely to imply their effectiveness from a global perspective.
With regard to transfer pricing regulation, the scientific literature also points to the effectiveness of this governmental restriction to limit profit shifting (Beuselinck et al., 2009; Lohse et al., 2012; Jost et a., 2010).
Jost et al. (2010) conducted interviews with managers being in charge of the transfer pricing operations of their multinational corporations. The authors found that managers perceive a high risk stemming from non-compliance with transfer price regulation since non-compliance results into substantial fines if detected by the authorities. This evidence hints that managers and firms recognize this regulation, and are impacted by the governmental transfer pricing regulation. Even though this study is qualitative in nature, it still provides a broad outlook of the effectiveness of transfer pricing regulations.
9
documentation requirements, where more extensive documentations causes higher effectiveness. In addition, the study by Lohse et al. (2012) highlighted that effectiveness also depends on the extent to which the implemented regulations are enforced within the legal system, where more sophisticated enforcement resulted into higher regulative effectiveness. Both studies demonstrate that differences in transfer pricing regulation exist, and that these differences determine the effectiveness of regulations, and also determine the extent to which corporations are impacted by it.
Given that these regulations to restrict profit shifting by multinational corporations are effective, it follows that there will be impacts on corporate behavior.
Buettner et al. (2012) already pointed to the impact of the firms’ capital structure. The authors demonstrated here that profit shifting regulation decreases the amount of internal debt funding, and leads multinational corporations to substitute this with external sources of capital.
Buettner et al. (2014) conducted a subsequent study which emphasizes on the resulting
implications for firms resulting from the implementing regulative measures by the government. The authors conclude that transfer pricing regulations affect the investment decisions of a multinational corporation, as well as their level of employment in the respective subsidiary. Further, evidence suggested that the implementation of such restrictions increases the tax-rate sensitivity of firms.
10
3. Hypotheses development
In the following part of this thesis, I will develop sound argumentation that both methods to restrict profit shifting, namely the thin-capitalization rules and the transfer pricing regulations, will have a negative impact on profit margins of multinational corporations. Further, I will discuss implications of the national tax rate which will be the degree of incentive for firms to engage in profit shifting initially. Sequentially, I will derive testable hypotheses which will be the center of the empirical analysis.
3.1 Thin-capitalization rules and profit margins
In order to demonstrate how the implementation or the increase of thin-capitalization rules will reduce the profit margins of multinational corporations, let us assume there exists a country without any capitalization restriction. For now, I will disregard impacts of the tax-rate, and discuss respective implications further below.
If firms in this country do not face any restriction on their financial structure, they are likely to use internal debt financing, for the reasons mentioned in the literature review. Given that debt payments are tax-deductible, firms will enjoy a reduced collective tax burden. These firms pay relatively low taxes, while simultaneously receiving revenues of comparable magnitude as related firms in the industry. If we assume that firms also have similar costs, this would imply that these firms ultimately have lower overall expenses which implies a higher profitability of these firms compared to domestics corporations which are unable to reduce their tax burden by profit shifting.
To highlight the negative implications on profit margins, we now assume the introduction of a thin-capitalization rule.
The introduction of this restriction will establish a threshold for the amount of allowable debt used by multinational corporations. All firms which rely extensively on internal debt as a means to finance their operations will now have excess debt exceeding the established threshold. Since this excess debt is not deductible anymore, the tax burden effectively increases, ceteris paribus. A higher tax burden will ultimately lead to increased costs for the multinational corporation. If we assume that revenues remain unchanged, the introduction of a thin-capitalization rule would effectively reduce corporate profit margins.
11
situation, and adjust their financial structure which would still result in reduced profit margins. This can be explained since the firm has to finance their structure in a less efficient way, and even the substitution with external debt would be more expensive.
As the example above theoretically demonstrated, I will assume that the implementation of thin-capitalization rules will ultimately lead to reduce profit margins, and therefore state the first hypothesis as:
H1: The increase in strictness of thin-capitalization rules will lead to lower profit margins of multinational corporations.
3.2 Transfer pricing regulation and profit margins
As I have done for thin-capitalization rules, I will demonstrate here theoretically how transfer pricing regulations will negatively impact corporate profit margins by using a similar theoretical example as above. After showing how transfer pricing regulation reduces profit margins, I will conclude with a testable hypothesis.
Again, I will start the assumption with a country which does not have any restrictions on transfer pricing, and I also disregard the national tax rate for now.
In the imagined country, firms price their internal corporate trades of goods and services in such a way that profits are minimized for the collective corporation. This would mean that if a firm would have other subsidiaries in countries with lower tax rates than this country, this firm would likely charge relatively low prices for exported goods and services, and thus realize relatively small amounts of profits in this country. This implies that the lower reported profits in this country would result in a lower tax burden for this subsidiary, and the remaining profit can be taxed elsewhere at lower corporate tax rates.
Contrary, multinational corporations for which the country’s tax rate is relatively low are likely to realize larger amounts of profits in this country. This implies that corporate profits are taxed at a lower rate which effectively reduces the overall tax burden for the collective corporation.
12
are allowed for intra-firm trade. The establishment of these arm’s length prices reduce the ability of firms to price their internal exchange of goods and services in such a manner that would most beneficial from a tax-reducing perspective. Therefore, the introduction forces multinational corporations to be less tax-efficient which ultimately increases the tax burden for the firm. Again, if we further assume that other firm characteristics such as the level of revenues does not change, the increase in taxes translates into higher corporate expenses. As a consequence, this implies a reduced profit margin for the restricted firm, ceteris paribus.
Thus, I formulate the following hypothesis:
H2: The increase in strictness of transfer pricing regulation will lead to lower profit margins of multinational corporations.
3.3 Tax rate sensitivity
The resulting impact on corporate profit margins does not solely depend on the presence or the increase of restrictive frameworks, but also on the respective statutory tax rate in a certain country. Again, I will demonstrate how tax rates interfere with each separate regulation of profit shifting, and conclude by establishing a hypothesis. In general, we assume that higher tax rates increase the likelihood of firms to engage in profit shifting since it becomes less profitable to realize profits in these countries (Buettner et al., 2014).
With regard to thin- capitalization rules, we assume that a country has implemented a certain degree of restriction, and theoretically demonstrate the implications of a change in the statutory tax rate in this imaginary country.
First, we assume the case of a firm which inhibits excess internal debt which implies that a certain amount of this debt is not tax-deductible as in the example above for thin-capitalization rules. Given that this excessive debt already triggers reduced profit margins for the multinational corporation, we further assume that this country would increase their statutory tax rate. This increase implies that the non-deductible debt which exceeds the threshold will be now taxed at a higher rate. Consequentially, this translates into higher tax burdens for these firms, and ultimately reduce their profit margins if other firm characteristics remain unchanged.
13
does not have excessive debt. Since all debt below this safe haven or safe harbor level is tax-deductible regardless of the applied statutory tax rate, these firms would not face any
consequences from an increase in the statutory tax rate. Even though the firm under this scenario are not affected by governmental restriction, the implementation or increase of the thin
capitalization rules will increase the likelihood for firms to be exposed to this restriction. To see this, assume that a country increases its threshold level which limits the allowed amount of debt used by the respective subsidiary. Given that the majority of corporations is financed by a certain amount of debt, the increased restriction would result into a larger amount of firms exceeding this new threshold. Given that excess debt reduces the profitability of a firm as argued above, an increase of thin-capitalization rules would still imply a negative impact on corporate profit margins since more firms will be exposed to restrictions.
Concluding, in either case, an increase in strictness of thin-capitalization rules would impact multinational corporations in a negative way, either on the individual basis with excess debt, or on the collective level with more firms exceeding the thresholds.
Regarding the impacts resulting from an increase of transfer pricing regulation, we again start our assumption with a country which has established a certain level of allowable debt. As I have outlined above, firms which are restricted in their ability to shift their profits, are forced to realize higher profits in this country. This implies that multinational corporations have more difficulties to avoid unfavorable taxation which ultimately leads to reduced profit margins. The resulting consequences for firms, and especially the degree of negative impact on a given firm depend now on the statutory tax rate in this country. Relating to the example, firms which perceive the
14
multinational corporations. As previously argued, increasing the statutory tax rate will intensify the adverse effects of these restrictions, and ultimately lead to additional reductions of profit margins. This reduced corporate performance manifests itself since both restrictions force firms to organize their operations in a less efficient way resulting from higher tax burdens. Given that higher tax rates increase the likelihood for firms to rely on profit shifting to increase profit margins, the presence of profit shifting restriction will have a more negative impact on corporations.
Therefore, I conclude with the following and final hypothesis:
H3: The negative impact on firms from profit shifting regulation will be more severe in high tax countries.
4. Methodology
In the following part of this thesis, I will outline the investigative approach which is taken to test my three hypotheses. After a brief discussion of the data underlying this study, I will discuss the main variables, and their reason for inclusion. Further, I will also discuss presumed relationships based on the theoretical argumentation in the previous section, and related findings in the
literature.
15
4.1 Dependent variable
In order to test the impact of increases in profit shifting regulation, and the influence of a nation’s statutory tax rate on the profitability of multinational corporations, I will employ the profit
margin of each firm as a respective measurement. This variable is the percentage of profit a corporation earns relative to its amount of sales. The nature of this variable makes any further adjustments of different firm sizes obsolete since it is already expressed in relative terms. The data for this variable is obtained from the Orbis database gathered by the Bureau van Dijk and spans the years 2005- 2009.
Based on the theoretical argumentations provided in the literature section, I expect that this variable decreases with increasing restrictiveness of profit shifting regulations, while controlling for influencing firm- and country-level characteristics which will be discussed below.
4.2 Independent variables
The two main independent variables in this regression are a proxy for the thin-capitalization rules, and a proxy for the degree of transfer pricing restrictions for countries on a national level.
4.2.1Thin-capitalization rules
With regard to thin-capitalization rules, this empirical analysis makes use of the ratio of allowed debt relative to equity which is restricted by the national tax law. This variable is presented by the units of debt which are allowed in relation to one unit of equity. For instance, the number three as a value for this variable would imply that the given country allows firms to have three times as much debt as equity. In other words, ¾ of the total firm’s assets can be financed by debt.
The data for this variable is taken from Buettner et al. (2014) which was originally gathered by the OECD for all member countries, and presented in the Appendix B: Countries (Table 5). In order to employ this variable, I use the transformation which has been suggested by Buettner et al. (2014) to increase the comparability of values among different countries.
The debt-to-equity value which I denote with (α) is transformed following Buettner et al. (2014), and the indicator is presented below:
Thin-capitalization rule= 1/(1+ (α)
16
restriction. This indicator is chosen since it limits any possible degree of restriction to a set range. Assuming the most restrictive case in which a nation does not allow the multinational firm to have any debt as means to financing (α= 0), the value of the indicator will be 1. Contrary, the case of no restriction of debt financing (α= ∞) would result in an indicator value of 0.
In order to illustrate the mechanism in more detail, I will refer to a numerical example of a country which has an established threshold of 3:1 (α= 3). Under this scenario, the respective indicator would have a value of 0.25 which results from employing alpha in the above equation. If we now assume that this country increases its thin-capitalization restriction to a debt-to-equity ratio of 2:1 (α= 2), and plug this number in the aforementioned equation, the value of the
indicator would be 0.33. Thus, increases in thin-capitalization rules will be represented by increases of the indicator’s value.
4.2.2 Transfer pricing regulation
With regard to transfer pricing regulation, I will use the classification system of Lohse et al. (2012) which establishes groups of countries being characterized by similar transfer pricing regulation. Their framework focuses on the documentation requirements, but also accounts for respective penalties and enforcement. The authors start be defining characteristics of transfer pricing regulation, and apply these to each country in the sample. In the case a certain criteria is met, the country receives the value of 1 for this feature, and the value of 0 if otherwise. Based on the sampled countries, Lohse et al. (2012) establish six different groups of transfer pricing regulation. According to the amount of positive characteristics which are satisfied by a given country, this country is then grouped with countries which obtain the same score on the tested criteria. Thus, the countries in a certain group all display similar characteristics of transfer pricing regulation. Further, the group 0 represents the lowest regulation of transfer pricing , and therefore includes countries with no implemented restrictions. The most restricted countries are grouped in class 5 which is the highest attainable group of regulative measures. The data for all the countries which are included in this study is included in the Appendix B: Countries (Table 5) for
17
Finally, I regard the implications resulting from the tax rate, or the changes thereof over the sampled period. In order to test the respective effects of the tax rate, I will include it as single variable, but also in interaction with the other two independent variables. This approach enables me on the one hand to identify the direct impacts on corporate profit margins, but on the other hand indirect impacts which result from reducing or enlarging the influence of either thin-capitalization rules or profit shifting regulations.
In order to proxy the national tax rate, I will employ the statutory corporate tax rate as a means to capture the level of taxation in a given country. This decision was mainly based on the conclusion derived by Clausing (2013), who studied the previous literature which employed effective, as well as statutory tax rates, or a both simultaneously (Bernard and Weiner, 1990; Hines and Rice, 1994; Collins et al., 1998). She concluded that both proxies achieve similar magnitude, but statutory tax rates achieve better statistical significance. Further, I claim statutory tax rates to be superior since they are restricted to a certain country, and cannot be influence by the
multinational corporation on the firm-level.
Based on the previous discussion of variables, I construct the following empirical model which will be sequentially employed to test the aforementioned hypotheses:
Profit margin= α1* tax rate + α2*(tax rate*profit shifting regulation) + α3* profit shifting regulation + control variables + ɛ,
where α1 is the coefficient of the statutory tax rate applied in a given country, α2 is the coefficient
for the interaction term and captures the moderating effect of the tax rate on the profit shifting regulations, α3 is the coefficient for the profit shifting regulation, control variables are a set of
18
4.3 Control variables
In order to control for the various influences which cause an impact in the dependent variable profit margin, I apply several control variables in order to isolate the impact of profit shifting regulations from other influencing factors which would disturb the empirical findings. These variables are included based on theoretical justification, and findings from previous scientific literature which provide scientific evidence for their impact.
4.3.1 Country-level controls
Due to the historical and geographical developments of a given country, each nation has its distinct economy, and therefore different external backgrounds which can potentially impact the profit margins of multinational corporations. Hence, I will employ country-level control variables which are intended to exclude these influences on profit margins, and thus account for
comparison among deviating fundamental economic conditions of each firm. All six country-level controls aim to eliminate the impact of economic conditions on profit margins being present in a certain country, and thus enable me to detect the isolated influence of the strictness of profit shifting regulation on corporate profit margins.
First, the literature presents evidence that the economic market of a country has an determining impact on the extent of a firm’s economic activities since it represents the firm’s ability to conduct its economic activities. Here, larger economies provide a larger potential to firms, since large economies have higher demand, increased buying power, and more sophisticated supporting industries which all fuel business activities. Demirguec- Kunt and Huizinga (1999) investigate the macroeconomic impact on corporate profitability, and especially focus on the level of GDP which is present in a certain country. The authors divided countries into different subgroups according to their level of GDP, and investigated the profitability of corporations which were located into these different countries. Empirical evidence pointed out that firms in countries with larger GDP were on average characterized by higher profitability. Thus, I control for effects resulting from the size of the market by using the proxy of total gross domestic product (GDP) which should depict the economic strength of a nation.
19
is also likely to increase corporate profitability since economies with increasing levels of GDP will sequentially have higher level of GDP which will imply higher profitability according to the authors. Therefore, I employ the rate of GDP growth as a percentage of relative changes in GDP. Third, previous literature identified labor productivity as an influencing force on corporate performance, and thus I control this impact by accounting for differences in the productivity of each inhabitant. Huizinga & Leaven (2008) investigate determinants of profit shifting activities by multinational corporations, and especially focus on country-level characteristics. The authors use earnings before interest and taxes as dependent variables and determine the influence of various country- and firm characteristics. Their empirical findings reveal a negative impact of labor productivity on the firm’s operating profits. Thus, as a representative measurement I include the GDP per capita in order to normalize this impact in the regression.
The last two potentially impacting variables account for different conditions in the financial market of each given country, and they are additional characteristics of a nation’s economy. Thus, I employ the inflation rate of each country, and further account for general economic differences by using the Indicator of economic freedom. The impact of the inflation rate is investigated by Liu (2004) who investigates how macroeconomic characteristics influence corporate failure. Lui (2004) finds that higher interest rates increase the rate of corporate failure and reduce firms’ profitability. Since interest rates are set in accordance to the level of a county’s inflation, this evidence implies that higher inflation, and thus higher interest rates, will lead to decreased profitability. In addition, the Index of economic freedom gains acceptance as a control variable from the study by Carlsson & Lundström (2002). Their study centered on the impact of economic freedom on economic growth. The authors obtained evidence that increased economic freedom has a positive influence on economic growth measured as a rise in GDP. Since higher economic freedom increases growth, and growth increases profitability, high degrees of
20
4.3.2 Firm-level controls
In addition to the above mentioned country-level control variables, I will also account for potential influences which stem from the characteristics of the firm itself.
The previous literature pointed out that the size of a firm has several impacts on the structural characteristics and on the activities of the corporation. Therefore, I employ the level of sales in order to control for these deviations in economic activity among firms. Corporate sales are used in this case instead of assets, since this study centers on the change in profits for corporations. Further, profit shifting regulations will impact the level of sales rather than the asset base of a firm, as it has been demonstrated by Grubert (2012).
5. Results
The following section of this thesis will center on the statistical analysis which is employed to test the aforementioned hypotheses. After a discussion of the descriptive statistics, I will focus on the correlations between variables, and finally discuss the outcome of the multiple regression analysis.
5.1 Descriptive statistics
The descriptive statistics of the employed variables are presented in the table below. This
21
Table1:
5.2 Correlations and outliers
As a sequential step before engaging in multiple regression analysis, I investigate the bivariate correlations among all variables in order to gain an initial insight of the relationships between the different employed variables, but more importantly to identify any correlation among
independent variables that might impact the regression, and induce measurement errors. In addition, I will investigate if the data inhibits any outliers which might impact further analysis. The correlation matrix overall displays statistically normal and acceptable correlations, but a few correlations among independent variables warrant further investigation.
The following table depicts the correlation matrix of the variables. Table 2:
Variables N Minimum Maximum Mean Standard deviation
Dependent varibale Profit margin 141565 -32.45 45.47 6.1414 10.36412 Independent variables Thin-capitalization 115025 1 8 2.9445 1.58742 Thin-capitalization index 144838 0.11 1 0.4392 0.3017 Tansfer pricing 144838 0 5 3.0486 .98696 Transfer pricing dummy 144838 0 1 0.4045 0.4907 Tax rate 144838 0.1 0.41 0.3159 0.05636 Tax rate dummy 144838 0.0 1.0 0.8271 0.3782
Controls
Year 144973 2005 2009 2007 1.1412 Sales 115508 0.00 617,144,794.30 820,156.54 6,335,967.83 Market size (GDP) 144838 26,085,307,221.85 13,681,138,648,037.80 2,089,420,375,581.57 2,798,364,013,684.88 Economic growth (GDP per capita) 144838 -14.74 14.16 1.1585 3.4031 Inflation rate 144838 -0.04 0.14 0.0222 0.0170 Index of economic freedom 144838 49.8 82.66 68.1923 6.7963
Year Profit margin Sales Thin-capitalization Thin-capitalization index Transfer pricing Tranfer pricing dummy Market size (GDP) Economic growth (GDP growth) Labor productivity (GPD per capita) Inflation rate Index of economic freedom Tax rate
Tax rate dummy Year 1 Profit margin -0.068** 1 Sales 0.012** 0.017** 1 Thin-capitalization 0.001 -0.027** -0.070** 1 Thin-capitalization index 0.000 0.057** -0.007* -0.951** 1 Transfer pricing 0.145** -0.019** 0.059** 0.004 -0.178** 1
Tranfer pricing dummy 0.040** -0.023** 0.056** 0.196** -0.229** 0.841** 1
Market size (GDP) 0.005* -0.006** 0.157** -0.384** -0.165** 0.307** 0.322** 1
Economic growth (GDP growth) -0.673** 0.101** -0.003 0.009** 0.178** -0.137** -0.095** -0.077** 1
Labor productivity (GPD per capita) 0.002 0.008** 0.049** -0.192** 0.057** -0.063** -0.117** 0.257** -0.221** 1
Inflation rate -0.136** 0.059** -0.005 0.044** 0.117** -0.019** -0.037** -0.064** 0.459** -0.356** 1
Index of economic freedom 0.075** -0.001 0.114** -0.269** -0.031** 0.068** -0.118** 0.420** -0.117** 0.567** -0.167** 1
Tax rate -0.221** -0.014** 0.072** -0.055** -0.399** 0.341** 0.343** 0.557** -0.014** 0.238** -0.225** 1
Tax rate dummy -0.146** -0.004** 0.019** -0.182** -0.260** 0.161** 0.061** 0.287** -0.064** 0.342** -0.132** 0.712** 1
22
Several correlations in the table warrant attention, and point out that the sample might inhibit multicollinearity. Therefore, I conduct a more detailed discussion of these problematic
relationships, and a test for multicollinearity in the Appendix. Overall, these results of these test indicate that the sample does not suffer from multicollinearity which enables me to proceed with the regression analysis.
Regarding outliers, I initially investigate the descriptive statistics above which hints that the variable Profit margin is likely to include several outliers due to their high minimum and maximum values. Additionally, the variable Sales also depicts extreme values which warrants further investigation. The remaining variables seems acceptable at first sight.
In order to detect the outliers in these variables, I transform the values into standardized z-scores, and compare them to the established statistical threshold of -2.5 and 2.5. Any z-score which has a value outside of this range is considered an outlier.
The variable Profit margin shows extreme standardized values on either end of the data which is below -2.5 on the lower end, and exceeds 2.5 on the upper end. Given the large amount of sampled firms, I decide to exclude either end of the data from the analysis without reducing the data by a large extend. This reduction of data reduces the sample from the original 147585 subsidiaries to 144838 which does not harm its representativeness.
Regarding the variable Sales, I again transform the original values of this variable into
23
5.3 Regression analysis
After the two previous sections have investigated the nature of the data, and further allowed me to proceed with the regression analysis, this section will now discuss the empirical findings, and further derive respective conclusions which focus on answering the theoretical hypotheses of this thesis.
The regression analysis is divided into six models which employ different combinations of variables. The respective results of each model are displayed in the table below, and will be discussed in chronological order. Overall, all models do not control for country- and year-fixed effects which enables me to derive comparisons across borders and across time. The control variables I employ in each model should provide sufficient characteristics of each country and each year so that additional controls for these two factors is obsolete. Further, the test for multicollinearity and the correlation matrix did not reveal severe impacts across years and countries which justifies to not separately control for these variables. Additionally, this approach fits the aim of this thesis and provides global results which should be mainly caused by the employed profit shifting regulation.
5.3.1 Model 1
24
Table 3:
The model achieves a fairly low R² of 0.137, but is nonetheless significant. All included variables enter the model significant at a 10% significance level or better with the exception of the index of economic freedom which inhibits a p-value of 0.160, and is thus not significant. In addition, the variable year slightly exceeds acceptable significance levels with a p-value of 0.116. Even though this result is surprising, it is likely that other variables already account for the majority of
economic characteristics represented by the index of economic freedom, so that the it becomes obsolete. Further, the same argument hold for the years, and points to the fact that the data does not suffer from any influences which depend on a specific time.
As expected and discussed in the methodology section, sales have a positive impact on profit margins which implies that larger firms are commonly characterized with larger profit margins. Even though this impact is statistically significant, the coefficient has such a low value that is unlikely to have any practical influences.
In addition, market size also enters as expected with a positive coefficient which implies that larger countries host firms that have larger profit margins. This could be explained since larger economies simply provide larger economic potential for the firm. Again, the tge size of the coefficient is not substantial which hints that it might not have practical influences on corporate profit margins. Further, the variable economic growth displays a positive relationship with profit margins which is in accordance with theoretical expectations. Following logical argumentation, countries which inhibit higher economic growth will host firms which grow on average about the same extent. If a firm has larger growth, it will also have increasing profit margins, relative to
Regression analysis Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Dependent variable Profit margin Profit margin Profit margin Profit margin Profit margin Profit margin
Constant 102.2776 * (60.9501) 268.0241 *** (-61.5468) 99.5668 (62.1203) 70.1123 (61.7260) 247.8660 *** (62.4264) -457.6539 *** (87.0614) Year -0.0476 (0.0303) -0.1313 *** (-0.0306) -0.0458 (0.0309) -0.0314 (0.0307) -0.1204 *** (0.0311) 0.2311 *** (0.0433) Sales 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) Market size (GDP) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) Economic growth (GDP growth) 0.3567 *** (0.0137) 0.2716 *** (0.0143) 0.3613 *** (0.0138) 0.3616 *** (0.0139) 0.2748 *** (0.0144) 0.1741 *** (0.0168) Labor productivity (GDP per capita) 0.0000 *** (0.000) 0.0000 (0.0000) 0.0000 *** (0.0000) 0.0000 *** (0.0000) 0.0000 (0.0000) 0.0000 (0.0000) Inflation rate 7.9799 *** (2.2495) 10.5876 *** (2.2548) 7.3333 *** (2.2606) 7.7270 *** (2.2557) 9.9368 *** (2.2690) 22.1734 *** (2.5002) Index of economic freedom 0.0091 (0.006) 0.0310 *** (0.0065) 0.0098 (0.0065) 0.0084 (0.0065) 0.0323 *** (0.0065) 0.0382 *** (0.0066) Tax rate -10.1046 *** (0.9840) -8.8920 *** (1.2394) -14.3594 (1.7912) -11.6605 *** (1.1475) -15.2483 *** (2.2377) -14.2673 *** (2.2380) Tax rate dummy 0.6291 *** (0.1220) 0.3682 *** (0.1266) 0.7026 (0.1243) 0.6671 *** (0.1255) 0.4018 *** (0.1279) 0.4454 *** (0.1279) Thin-capitalization index not included -2.8805 *** (0.7121) not included not included -3.8042 *** (0.7708) -3.8605 *** (0.7708) Thin-capitalization index * Tax rate not included 17.8738 *** (2.5090) not included not included 21.0288 *** (2.7077) 21.7127 *** (2.7067) Tranfer pricing not included not included -.302 *** (0.1459) not included 1.7419 *** (0.5117) -0.3301 *** (0.1582) Tramsfer pricing * Tax rate not included not included 1.214 *** (0.4776) not included -0.5153 *** (0.1575) 1.1059 *** (0.5143) Transfer pricing dummy not included not included not included -1.275 *** (0.3877) not included
Transfer pricing dummy * Tax rate not included not included not included 3.919 *** (1.2003) not included
Financial crisis dummy -1.662 *** (0.1430)
N 144838 144838 144838 144838 144838 144838
R2 0.137 *** 0.149 *** 0.137 *** 0.137 *** 0.150 *** 0.153 ***
25
firms experiencing smaller growth.
The variable labor productivity also impacts firms’ profit margins in a positive manner, and the same argumentation as for market size can be employed since labor productivity is measured as GDP per capita. Thus, higher labor productivity leads to more efficient and effective labor, and thus results into larger profit margins for firms.
Moreover, the variable Inflation rate displays a large and positive coefficient which is statistically significant. This is contrary to my previous theoretical argumentation, but can be explained when we include the relative value of money. I argued that high inflations rates will have a negative impact on profit margins since they indicate economic instability. Even though this instability will impact profit margins negatively, I neglected to consider the value of money in relation to foreign currencies. Further, high levels of inflation lead to a decrease in the domestic currency which implies that foreign currencies relatively gain in value. Since foreign currencies increase, it is relatively cheaper for foreign firm to import from this country which sequentially fuels
respective sales and profit margins (Mann, 1986). Therefore, these positive impacts offset the negative effects of economic instability, and ultimately lead to increased profit margins of firms being located in high inflation countries.
Additionally, the indicator of economic freedom is found to be positively related to profit margins which is in accordance to previous expectations. I argued that countries with higher economic freedom are likely to have higher economic capabilities which would result in higher profit margins. Still, this coefficient does not achieve statistical significance which restricts any further conclusions.
26
5.3.3 Model 2
The second model resembles the previous model, but additionally includes the variables Thin- capitalization index and the interaction term of this index with the tax rate (Table 3). The model achieves an increased R² in comparison to the first model, and explains 0.149 of the dependent variable’s variance. The model displays a few changes in p-values, but the coefficients of the control variables maintain their directional impacts in general.
Three changes in this model are striking though, and I briefly discuss each in the following. In contrast to the first model, the year variable achieves statistical significance now, and displays a negative coefficient. Thus, the more recent year are characterized by lower profit margins of multinational corporations. This finding might be due to the global financial crisis in 2007, and I will further investigate this impact in the final model of this regression analysis.
The second change in this model is that the variable labor productivity now lacks statistical significance. Given the previously detected low coefficient which indicated solely a minor impact on corporate profit margins, this insignificance is not alarming.
The two newly added variables also enter the model significantly, and achieve p-values below the 1% significant level. The variable which measures the strictness of thin- capitalization regulation depicts a fairly high and negative coefficient, as I theoretically assumed. This finding implies that thin- capitalization regulations are indeed effective in restricting profit shifting of multinational corporations due to the negative impact obtained by this regression. Additionally, increasing or tightening this regulation will reduce corporate profit margins, and thus hurt the profitability of multinational corporations.
27
high tax countries. These findings provide evidence for the first hypothesis, and further partly confirm the third hypothesis.
5.3.3 Model 3
The third model resembles again the first model, and includes all control variables and profit margin is the dependent variable. Similar as in the second model, the p-values of the control variables experience slight changes, but maintain their directional impact overall (Table 3). The only change among control variables is that the year again like in Model 1 does not enter the model significantly. Further, labor productivity now achieve statistical significance but only inhibits a small coefficient, and the variable Index of economic freedom losses its previous significance. The model further includes the transfer pricing classification by Lohse et al. (2012) as a new variable, and also an interaction term of this variable and the tax rate is added to the model. These variables are used in order to investigate the second method of profit shifting on corporate profit margins, and the model achieves a significant R² of 0.137 which is lower than the previous Model 2.
In accordance to theoretical expectations, the coefficient for transfer pricing depicts a negative relationship which implies a reduction in profit margins, but is not as strong as the previous negative impact of thin- capitalization regulation. This implies that transfer pricing regulation does effectively limit profit shifting by multinational corporations, and sequentially reduces corporate profit margins.
The additional interaction term displays a positive coefficient which indicates that this negative impact of transfer pricing regulation is increased for countries characterized by higher tax rates.
5.3.4 Model 4
28
relatively strict transfer pricing regulation. I establish a dummy variable which is unity if a country is classified as group 4 or higher, and all other groups take the value of 0 which implies lower restrictive transfer pricing regulation. This approach thus broadens the categories of transfer pricing regulation which might increase the strengths of the respective coefficients, and also achieve increased statistical significance which providing robustness to the conclusion of the third model above.
Therefore, the fourth model resembles the third model, but uses the previously discussed dummy variable and a respective interaction term with the tax rate instead of the original classification system by Lohse et al. (2012). Again, profit margin is the dependent variable, and all
aforementioned control variables are included.
As in previous models, Model 4 experiences some minor deviations among p-values for the control variables, but achieves an overall significant model fit. The predictive R² value remains the same as in Model 3 and has a value of 0.137 which I expected due to the high similarity between these two models.
The two newly added variables which proxy the strictness of transfer pricing regulation enter the model as theoretically expected. The dummy variable of transfer pricing has a negative
coefficient which would imply that increased regulations would reduce corporate profit margins which also aligns with the original measurement of transfer pricing regulation in the previous model. Further, the coefficient is of larger size which implies a stronger impact on profit margins, and hints that the employed approach improved the representation of this relationship among variables.
In addition, the interaction term of the dummy variable and the tax rate depicts a positive
coefficient which hints that higher tax rates increase the negative effect of increasing the transfer pricing regulation in a certain country. Similar as the dummy variable itself, this interaction term displays a stronger relationship with profit margins which implies that the implemented approach is better suited relative to the previous model.
Overall, these finding point out that transfer pricing regulation is an effective means by
29
5.3.5 Model 5
The fifth model combines Model 2 and Model 4, and thus includes the Thin- capitalization index variable, and also the Transfer pricing regulation variable (Table 3). Additionally, the respective interaction terms for the aforementioned variables are added, and all previous control variables complete the model which again uses the profit margins as the dependent variable.
The constructed model is significant with an R² of 0.150 which makes it the most predictive model so far. The model inhibits some changes in statistical significance among control variables, where the most noticeable variable is the insignificance of labor productivity. This insignificance is not alarming, and it is likely that part of the variable’s explanatory impact is captured by the market size which already includes a country’s value of GDP. All remaining variables enter the model with similar coefficients as before, and achieve statistical significance at the 1%-level. The most important variables, which represent the two techniques to regulate profit shifting, enter the model in the same manner as they did when previously included in isolation. Thus, thin- capitalization remains significant, and impacts corporate profit margins negatively. The
interaction term of this variable and the tax rate displays a strong and positive coefficient which translates into an increase of the negative relationship between thin- capitalization regulation and profit margins. Further, the proxy for transfer pricing regulation displays the theoretically correct relationship, and also achieves statistical significance. The same holds true for its respective interaction term which displays the expected positive coefficient, and enters the model significantly.
30
shifting regulation has more adverse effects in countries with high corporate tax rates. Rephrasing the aforementioned, higher tax rates increase the negative impact of profit shifting regulation on profit margins for both types of profit shifting regulation.
5.3.6 Model 6
This final model resembles the previous Model 5, but inhibits an additional dummy variable which should capture the impact of the recent financial crisis on the profit margins of
multinational corporations (Table 3). As Model 1 indicated, multinational corporations inhibited lower profitability in the more recent years. These lower profit margins might be caused by the impact of the recent financial crisis, and thus might additionally impact corporate profitability. In order to investigate this impact and detect further impacts on profitability, I separate the sample in two different groups based on year of the crisis. The first group includes the years 2005-2007, and represent multinational corporations and their profitability before the financial crisis, and the second group includes the years 2008-2009 which represent the same corporations after the crisis. If the profit margin of a subsidiary is after the financial crisis, this dummy takes the value of 1, and the value of 0 otherwise.
The obtained model achieves statistical significance, and has an overall predictive ability of corporate profitability of R²= 0.153. The added dummy variable enters with a negative coefficient which is significant at the 1%-level, and the remaining variables maintain similar coefficients and p-values as in the fifth model. The significant negative coefficient of the dummy variable lets me infer that the crisis indeed had an negative impact on the profitability of
31
6. Conclusion
This thesis has centered on the issue of profit shifting regulation, and its respective impacts on the profitability of multinational corporations. I focused on the two dominant techniques which are employed by firms to shift their profits internationally, namely the intercompany debt financing and the transfer pricing of intra-firm trade. More precisely, I discussed how multinational
corporations reduce the overall corporate tax burden by implementing these techniques, and thus increase their profitability. I pointed to the competitive advantage of profit shifting for
multinational corporations which arises since their domestic rivals are unable to lower their tax burden by shifting their profits to lower tax jurisdictions.
Governments around the globe thus aim to restrict these practices by multinational corporations in order to sustain their tax revenues, and achieve fair and equal competition among firms. Sequentially, I reviewed the two major restrictive measurements which are taken by government to combat profit shifting, and further explained their underlying mechanisms. In the following, I discussed how these regulations theoretically are intended to influence corporate profit margins, and derived hypotheses in order to test the effectiveness of these regulations, and identify their impact of corporate profitability.
The empirical evidence of this statistical investigation pointed out that stricter regulation of debt financing (thin- capitalization regulation) was found unfavorable for corporation since it led to lower profit margins, and thus reduced their profitability. In addition, findings hinted that this adverse effect further increases if the country is characterized by higher corporate tax rates. In accordance with theoretical argumentation, the same negative impact was also proven for transfer pricing regulation. Here, statistical evidence supported the effectiveness of this regulation, and further confirmed a negative impact on corporate profit margins. Thus, thin- capitalization regulation and transfer pricing restriction both limit profit shifting by multinational corporation effectively, and thin- capitalization regulation was found to be the more effective technique.
32
studies should replicate this research with different dataset, including additional countries, and employing larger timeframes. Thus, following research should include a measurement of the actual level of profit shifting of each subsidiary, or employ an aggregate measurement of the level of profit shifting which is inherent in each country. This would achieve richer insights on the effectiveness of both restrictive profit shifting regulations, and also identify their individual impact and their extent to reduce corporate profitability. Further, it would be possible that additional studies investigate countries which experienced a change in profit shifting regulation, and compare these finding to a control group of countries lacking any regulative changes.
Moreover, any replication of this thesis which employs different proxies, other control variables, or a new dataset will improve the robustness of this study.
6.1 Implications
This study has several practical implications for political decision-making, but also for managers of multinational corporations.
With regard to governmental regulation, policy makers should be aware that the restriction of profit shifting not only creates a fair economic environment for domestic firms and foreign subsidiaries in a country, since it restricts the ability of multinational corporations to reduce their tax burden, but also features additional implications for multinational corporations. Even though fair competition is beneficial for the national economy and its domestic firms, governments should not disregard the reduction in profitability of multinational corporations. By increasing the profit shifting restrictions, governments create a disadvantage for these foreign subsidiaries hosted in the country when in comparison with their rivals on the international level. These multinational corporation which are located in the restricted country are likely to have a
disadvantage towards other multinational corporations which operate in less restricted countries, and which are able to reduce their tax liability. This limitation might lead to a reduction of the nation’s economic attractiveness, and sequentially harm the country’s foreign investments inflows which might further translate into lower tax revenues from hosted multinational
corporations. This negative impact should be taken into account when implementing restriction of corporate profit shifting, and especially in countries which inhibit higher tax rates since the adverse effects of regulation will have an even larger impact.
33
regulation in their foreign investment decisions since their respective profitability is influenced partly by the extent to which profits can be shifted. Therefore, identical project in different countries should are rather realized in countries with lower tax rates and less restricted profit shifting since these characteristics would enable the multinational corporation to achieve a higher profitability in comparison to countries characterized by higher tax rates and stricter profit
34
7. References
Bartelsman, E., Beetsma, R. 2003. Why pay more? Corporate tax avoidance through transfer pricing. Journal of Public Economics 87, 2225–2252.
Bernard, J.-T., Weiner, R.J. 1990. Multinational corporations, transfer prices, and taxes: evidence from the US petroleum industry. In Taxation in the global economy (pp. 123-160). University of Chicago Press.
Beuselinck, C., Deloof, M., Vanstraelen, A. 2009. Multinational income shifting, tax enforcement and firm value. Unpublished working paper, Tilburg University.
Bucovetsky, S., Hauer, A. 2008. Tax competition when firms choose their organizational form: should tax loopholes for multinationals be closed? Journal of International
Economics 74, 188-201.
Buettner, T., Overesch, M., Schreiber, U., Wamser, G. 2012. The impact of thin-capitalization rules on the capital structure of multinational firms. Journal of Public Economics 96, 930-938.
Buettner, T., Overesch, M., Wamser, G. 2014. Anti Profit-Shifting Rules and Foreign Direct Investment. Unpublished working paper. Center for Economic Studies & Ifo Institute, Ludwig- Maximilians University, Munich.
Carlsson, F., Lundström, S. 2002. Economic freedom and growth: Decomposing the effects. Public choice 112(3-4), 335-344.
35
Collins, J., Kemsley, D., Lang, M. 1998. Cross-jurisdictional income shifting and earnings valuation. Journal of Accounting Research 36, 209–230.
Demirgüç-Kunt, A., Huizinga, H. 1999. Determinants of Commercial Bank interest margins and profitability: Some international evidence. World Bank Economic Review 13 (2), 379-408
Demirgüç-Kunt, A., Huizinga, H. 2001. The taxation of domestic and foreign banking. Journal of Public Economics 79, 429–453.
Desai, M.A., Foley, C.F., Hines, J.R. 2004. A multinational perspective on capital structure choice and internal capital markets. Journal of Finance 59, 2451-2487.
Dharmapala, D., Hines, J. R. 2009. Which countries become tax havens?. Journal of Public Economics 93(9), 1058-1068.
Grubert, H. 2012. Foreign taxes and the growing share of US multinational company income abroad: Profits, not sales, are being globalized. National Tax Journal 65(2), 247-275.
Grubert, H., Mutti, J. 1991. Taxes, tariffs and transfer pricing in multinational corporate decision making. Review of Economics and Statistics 73, 285–293.
Hall, M., Weiss, L. 1967. Firm size and profitability. The Review of Economics and Statistics, 319-331.
36
Hong, Q., Smart, M. (2010). In praise of tax havens: International tax planning and foreign direct investment. European Economic Review 54(1), 82-95.
Huizinga, H., Laeven, L. (2008). International profit shifting within multinationals: A multi-country perspective. Journal of Public Economics 92(5), 1164-1182.
Jost, S.P., Pfaffermayer, M., Winner, H. 2010. Transfer pricing as a tax compliance risk. Unpublished working paper. University of Salzburg, Salzburg.
Liu, J. 2004. Macroeconomic determinants of corporate failures: evidence from the UK. Applied Economics 36 (9), 939-945
Lohse, T., Riedel, N., Spengel, C. 2012. The increasing importance of transfer pricing regulations–a worldwide overview. Unpublished working paper. Oxford Center for Business Taxation, Oxford
Mann, C. L. 1986. Prices, profit margins, and exchange rates. Federal Reserve Bulletin 72, 366.
Mintz, J., Smart, M. 2004. Income shifting, investment, and tax competition: theory and evidence from provincial taxation in Canada. Journal of Public Economics 88, 1149-1168.
OECD. 1987. Thin-capitalization and taxation of entertainers, artistes and sportsmen, Committee on Fiscal Affairs, Issues in International Taxation No. 2, Paris.
OECD. 2010. Transfer pricing guidelines for multinational enterprises and tax administrations, Paris.
