• No results found

The Corporate Tax Effect on Inflows of Foreign Direct Investment: The case of OECD countries

N/A
N/A
Protected

Academic year: 2021

Share "The Corporate Tax Effect on Inflows of Foreign Direct Investment: The case of OECD countries"

Copied!
36
0
0

Loading.... (view fulltext now)

Full text

(1)

J

Ö N K Ö P I N G

I

N T E R N A T I O N A L

B

U S I N E S S

S

C H O O L

JÖNKÖPING UNIVERSITY

The Corporate Tax Effect on Inflows of Foreign Direct

Investment

The case of OECD countries

Bachelor Thesis in Economics Author: Mihai Schendra

Aleksandar Zahariev

Tutor: Ph.D. Candidate Sofia Wixe Prof. Lars Pettersson

(2)

Bachelor’s Thesis within Economics

Title: The Corporate Tax Effect on Inflows of Foreign Direct Investment (The case of OECD Countries)

Author: Mihai Schendra, Aleksandar Zahariev

Tutor: Ph.D. Candidate Sofia Wixe, Prof. Lars Pettersson Date: May 2011

Keywords: tax, corporate income tax, foreign direct investment

Abstract

There is a reasonable amount of literature and discussions among scholars on the effect of host country corporate taxation on the inflows of foreign direct investment (FDI). This study is an attempt to analyse this effect in 25 high income OECD countries over the period of 1996 until 2009. The main objective of the paper is to prove statistically whether there is significant and negative relationship between the inflows of FDI and corporate taxation in the selected sample of OECD countries during the specified time span. This relationship is investigated with OLS regression analysis with pooled panel data to find to what extent the selected explanatory variable effective tax rate (ETR) along with trade openness, long term interest rate, share of internet users and labour cost have an impact on the dependent variable - FDI relative to GDP. Finally, it is proved that the elasticity between corporate taxation and FDI is positive at a level below the average effective tax rate and negative above the average level of effective tax rate. In addition, all other important variables included in the regression model are found to be significant determinants of FDI. The study is based on relevant literature and the statistical analysis is made in regard to the models described in scientific articles in the paper.

(3)

Table of Contents

1

Introduction ... 1

1.1 Problem Discussion and Purpose ... 2

1.2 Method ... 2

1.3 Limitations ... 2

1.4 Outline ... 3

2

Background ... 4

2.1 FDI Effects Upon Economies ... 4

2.2 Types of Taxes ... 5

2.3 Taxation of FDI ... 6

2.4 Tax Effect on FDI ... 7

3

Theoretical Framework ... 9

3.1 The Eclectic Theory of International Production ... 9

3.2 Previous Studies ... 9

3.2.1 David Hartman (1985) ... 9

3.2.2 Michael Boskin and William Gale (1987) ... 10

3.2.3 Joel Slemrod (1990) ... 11

4

Empirical Analysis ... 12

4.1 Data and Methodology ... 12

4.2 The Dependent Variable ... 12

4.3 The Determinants ... 13

4.4 Relationship Between the Determinants ... 16

4.5 Functional Form ... 16

4.6 Regression Model and Results ... 18

5

Conclusions ... 22

5.1 Suggestions for Further Research ... 22

Acknowledgements... 23

References ... 23

Data Sources ... 24

(4)

List of Figures

Figure 1 Average ETR in % in 25 OECD Countries for 1996 - 2009 ... 1

Figure 2 FDI Components ... 3

Figure 3 Paper Outline ... 3

Figure 4 Average GDP in current MLN US $ in 25 OECD countries for 1996 util 2009 ... 4

Figure 5 Average FDI in current MLN US in 25 OECD Countries for 1996 until 2009 . 4 Figure 6 OLI Paradigm ... 9

Figure 7 Log(ETR) to Log(FDI/GDP) ... 17

Figure 8 Histogram of Effective Tax ... 25

Figure 9 Histogram of Effective Tax below the average ETR (30%) ... 25

Figure 10 Histogram of Effective Tax above the average ETR (30%)... 25

Figure 11 Histogram of Standardized Residuals ... 26

Figure 12 Model 1 ... 27

Figure 13 Model 1 Autoregressive... 27

Figure 14 Model 1 Autoregresed Test for Autocorrelation ... 28

Figure 15 Model 1 Initial test for Autocorrelation ... 28

Figure 16 Model 2 ... 29

Figure 17 Model 2 Autoregressive... 29

Figure 18 Model 2 Initial Test for Autocorrelation ... 30

Figure 19 Model 2 Autoregressed Test for Autocorrelation ... 30

Figure 20 Model Effective tax <30 % ... 31

Figure 21 Model Effective tax >30 % ... 31

List of Tables

Table 1 Types of Corporate Tax Interpretation ... 6

Table 2 Correlation Matrix ... 16

Table 3 Regression Output... 19

Table 4 List of Countries ... 24

Table 5 Statistical Parameters ... 26

List of Equations

Equation 1 Effective Tax Rate Formula ... 5

Equation 2 Dependent Variable Formula ... 12

Equation 3 Share of Internet Users Formula... 14

Equation 4 Trade Openness Formula ... 14

Equation 5 Formula for Model I ... 18

(5)

Abbreviations

AR CIT DIA DUM ETR FDI GDP GNP IMF LC MNE M&A OECD OLI OLS R STR TO VAT Autoregressive

Corporate Income Tax Direct Investment Abroad Dummy Variable

Effective Tax rate

Foreign Direct Investment Gross Domestic Product Gross National Product International Monetary Fund Labor Cost

Multinational Enterprise Merger and Acquisition

Organization for Economic Co-operation and Development Organization, Location and Internalization

Ordinary Least Squares

Long Term Risk -Free Interest Rate ( 10 year yield government bond) Statutory Tax Rate

Trade Openness Value Added Tax

(6)

1

Introduction

Foreign Direct Investment (FDI) has been regarded as a key factor for economic growth and productivity for most economies and it is a highly discussed topic in the field of economics and finance. It is also considered a vital part of the rapidly evolving process of international economic integration. According to OECD Factbook (2010) FDI creates direct, stable and long-lasting links between economies and encourages the transfer of technology and know-how between countries. Moreover, FDI is an additional source of funding for investment and, under the right policy environment it can be an important vehicle for enterprise development (OECD Factbook 2010).

It is of general interest whether the tax cuts have a positive effect on government revenue. Although it is usually assumed that lower tax rates implies lower tax revenues, it has to be noted that the tax cuts have an indecisive effect upon the government budget. This is because, on one hand, lower tax can lower the tax collection but on the other hand lower tax would increase the inflows of investments in the host country and thus more tax would be collected from the new investments.

Because of the importance of inflows of FDI for economies governments are constantly decreasing their tax rates on corporate profits, also known as tax competition, in order to attract more inflows of foreign sourced income. Devereux (2008) observes this phenomenon in developed economies and according to his estimates corporate taxes have dropped substantially since the early 1980s. The average rate in OECD member states in the beginning of the 1980s was nearly 50%; by 2001 it had fallen to under 35% (Devereux, Lockwood and Redoano 2008). Using data1 from OECD tax database to plot Figure 1 from below, it can be observed that the average effective tax rate (ETR) for 25 OECD countries has fallen from 36% to 26% for the period between 1996 and 2009.

Figure 1 Average ETR in % in 25 OECD Countries for 1996 - 2009

1 OECD Tax Database. Retrieved at 15th of April 2011from www.oecd.org/ctp/taxdatabase 36.1 35.3 34.3 33.6 32.9 32.0 31.5 30.9 29.6 28.6 28.2 27.7 26.6 26.5 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 A ve rag e E ff e ctiv e Tax i n %

(7)

It is widely believed that the host country taxation is one of the most important determinants for foreign investors when it comes to location decisions. According to Easson (2004) tax is

not an important factor in the initial decision to invest abroad. However, it has been the focus

of most empirical studies, since it is closely related to the amount of profit that is available for distribution (Easson 2004). Moreover, it is believed by many scholars that corporate income

tax (CIT) is a very powerful instrument in attracting inflows of capital but, at the same time, it

might have negative effect upon the tax revenue.

1.1

Problem Discussion and Purpose

The underlying purpose of the paper is to test statistically if there is a negative and at the same time significant relationship between inflows of FDI and the ETR in the selected sample2 of OECD countries for the time period between 1996 until 2009. This question has been the subject of many studies over the past 30 years and the answers differ widely (OECD, Corporate Tax Incentives for Foregin Direct Investment 2001).

1.2

Method

This paper is written with information from scientific journals, papers and data from OECD, World Bank, UNCTAD and IMF suited best for our research. Moreover databases, search engines on the Internet e.g. Julia and J-store and books have been used. The methodology used in this analysis is quantitative research and the relationship between the variables is calculated using the Ordinary Least Squares (OLS) estimates for pooled panel data.

1.3

Limitations

As every study this paper has limitations. Firstly, the study may not be consistent enough due to bias in the econometric model - the time span of the study is for a period of 14 years (1996 until 2009) due to unavailability of data for the chosen variables. Moreover, only 25 countries are selected due to the same reason – constraints on data for the particular model. Therefore, the model explains only the recent relationship between corporate tax and FDI.

Besides the temporal issue of the model there might be considerations regarding the sample countries used. The model includes only 25 developed economies, therefore feasibly the model that would apply to developed countries would not apply to developing economies. Another very important aspect of the study is that statistical information on FDI involves financial flows that do not necessarily correspond to the allocation of real investment, also called Greenfield investment. Actually, FDI is comprised of several types of capital. First it contains real investment (Greenfield) in plants and equipment. Second, a major part of FDI consists of the financial flows associated with mergers and acquisitions (M&As). Thus, this implies an ownership change in the absence of any real investment (De Mooij and Ederveen 2003). In other words, M&A involves transfer of ownership from domestic to foreign hands,

2 Australia, Austria, Belgium, Canada, Czech Republic, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Japan, Mexico, Netherland, New Zealand, Norway, Poland, Portugal, Slovak Republic, Spain, Sweden, Switzerland, United Kingdom

(8)

but do not create new productive facilities. OECD estimates (OECD, Recent Trends in Foreign Direct Investment 2000) suggest that M&As account for more than 60% of all FDI in developed countries. Hence, distinction between the different types of FDI is crucial because the different components of FDI may respond differently to taxes.

Figure 2 FDI Components

1.4

Outline

In order to understand the relationship between tax and FDI the next part - section 2 continues with the importance of FDI, followed by tax terminology and theoretical explanation of this specific relationship. Later the paper is split into literature review and quantitative analysis of the topic. Section 3 continues the theoretical framework of the study. Section 4 presents the quantitative part, where the theories analysed will be used as the foundation for the econometric models. The final model will be selected according to goodness of fit and coefficient significance. Lastly we will verify or reject our hypothesis through an analysis and conclude with further research topics in section 5.

Figure 3 Paper Outline

40%

60% Greenfield Investment M&A

Section 2. Background

•FDI Effects upon Economies •Tax Terminology

•STR and ETR •Personal Income Tax

and Other Taxes •Transfer Pricing •Taxation of FDI •Tax effect on FDI

Section 3. Theoretical Framework

•The Eclectic Theory by Dunning (1981) •Prvious Studies

•David Hartman (1985)

•Michael Boskin and William Gale (1987) •Joel Slemrod (1990) Section 4. Empirical Analysis •Data and Methodology •The Dependent Variable •The Determinants •Relationship Between the Determinants •Functional Form •Regression Model and

Results •Analysis Section 5. Conclusions •Final Remarks •Suggestions for Further Examinations

(9)

2

Background

2.1

FDI Effects Upon Economies

“FDI reflects the objective of establishing long term relationship by a resident enterprise in one economy (direct investor) in an enterprise (direct investment enterprise) that is resident in an economy other than that of the direct investor. This relationship is represented by direct or indirect ownership of 10% or more of the voting power of an enterprise resident in one economy by an investor resident in another economy.” (OECD, OECD Benchmark

Definition of Foreign Direct Investment 2008)

Foreign Direct Investment (FDI) has been regarded as a key factor for economic growth and productivity for most economies and it is a highly discussed topic in the field of economics and finance. FDI involves the transfer of financial capital, technology and skills. This process raises the question of the costs and benefits involved in it for the home and host countries and it is not clear, what costs are borne and what benefits are received by both sides (Moosa 2002). From a cost-benefit perspective one of the most important aspects of FDI is its effect on output and therefore on growth in the host country. As it can be seen from Figure 3 the average levels of GDP for the selected sample of countries has been rapidly increasing for the period of 2001 to 2008. It can also be observed from Figure 4 that FDI has been increasing for the same period except the years from 2001 to 2004. Thus, following economic theory and the plotted figures it can be logically assumed that GDP moves the same direction as FDI and FDI contributes to that growth to some extent.

Figure 4 Average GDP in current MLN US $ in 25 OECD countries for 1996 util 20093

Figure 5 Average FDI in current MLN US in 25 OECD Countries for 1996 until 20094

3 World Bank World Development Indicators. Retrieved at 15th March 2011 from http://data.worldbank.org 4 OECD International Direct Investment Statistics. Retrieved at 10th March 2011 from http://stats.oecd.org/

500000 600000 700000 800000 900000 1000000 1100000 1200000 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 0 10000 20000 30000 40000 50000 60000 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009

(10)

2.2

Types of Taxes

In any developed economy every single person is paying taxes and every government depends on the revenue collected from the taxes paid by the public. According to Oxford online dictionarytax is defined as:

“A compulsory contribution to state revenue, levied by the government on workers' income and business profits, or added to the cost of some goods, services, and transactions.”5

There are many different types of taxes and the terminology associated with it is very complex. Taxes on corporate income can be characterized as flat - constant rate is charged for all income levels; progressive - tax increases as income increases, or regressive - tax increases as income decreases. Taxes can be also classified as direct and indirect. Direct tax, as stated in Oxford online dictionary is:

“A tax, which is levied on the income or profits of the person who pays it, rather than on goods or services” and indirect tax is “a tax levied on goods and services rather than on income or profits.”

CIT, personal, withholding, inheritance tax belong to the first category, while value added tax (VAT), tax on cigarettes, alcohol and fuel belongs to the latter. Having an idea about the tax terminology is crucial for deeper understanding of the topic. In this paper we are mainly interested in the CIT and especially the ETR.

STR and ETR

According to Easson (2004) the tax that most foreign investors look for is the tax on business profits and it has been the focus of most of the empirical studies because it affects the amount of profit that is available for distribution. In other words, taxes have a negative impact on cash flows and profits, i.e. they are relevant for the determination of post-tax results. What matters mostly to the potential investors is the total amount of tax payable - the ETR and not the the statutory tax rate. Table 1 presents description of different interpretations of corporate taxation.A lot of researchers use STR for investigating the relationship between tax and FDI. However, the tax treatment of FDI is generally a complex issue. In reality STR is the maximum rate a corporation can pay. Corporate tax rules allow deductions and exclusions and that is why the STR is a poor guide to the actual disincentives the corporate taxation creates. Therefore, using STR can be misleading. It is impossible to capture all the complex details of the tax system that potentially affect foreign investment in an empirical analysis. Therefore, most studies rely on some type of ETR. An ETR is a rough proxy variable that summarizes the interaction of various tax rules on an investment (De Mooij and Ederveen 2003). It can be expressed by the following formula:

Equation 1: Effective Tax Rate Formula

(11)

Table 1: Types of Corporate Tax Interpretation

Corporate Income Tax (CIT) Description

Effective Tax Rate (ETR)

The actual tax paid divided by net taxable income before taxes expressed as a percentage. It is different from the STR because there are different tax rules that allow deductions and exclusions and ETR summarizes all of them. ETR represents the share levied on distribution of domestic source income to a resident individual shareholder, taking account of corporate income tax, personal income tax and any type of integration or relief to reduce double taxation. It is a more accurate measure because it considers the tax paid by companies as compared to the STR that does not consider the tax incentives undertaken by governments.

Statutory Tax Rate (STR)

The rate that is imposed on taxable income of corporations, which is equal to corporate receipts less deductions for la-bor costs, materials, and depreciation of capital assets. It is the maximum tax that a corporation can pay in principle6.

Personal Income Taxes and Other Taxes

Whether paid directly by the enterprise or paid by employees, all taxes (personal income taxes, payroll taxes, social security contributions, consumption taxes, custom duties and import taxes) are of concern to foreign investors primarily because of their effect on labor costs. It is believed that consumption taxes such as VAT and excise duties are of relatively little concern to foreign investors. In theory, those taxes are borne by consumers, rather than businesses (Easson 2004). Customs duties are of minor importance in developed economies. In developing countries, by contrast, they are often one of the most important sources of government revenue and are imposed at relatively high rates.

2.3

Taxation of FDI

In many economies the return on FDI may be subject to double taxation. Every foreign subsidiary is always subject to CIT in the host country. The realized profits can be taxed again under the marginal CIT in the home country of the parent company. Because this might discourage conducting international business, most economies avoid it by means of bilateral tax treaties based on the OECD Model Tax Convention (De Mooij and Ederveen 2003). For the sake of resolving the problem of interpretation and application of the taxation laws it was required to construct a legislation that addresses tax issues. Since 1992, OECD has been publishing the “Model Tax Convention on Income and Capital”. The document is a two volume paper which presents tax laws and commentary for OECD countries and 16 non-member countries. The part that is of interest to this study is Article 7 published on each edition of the paper. This article addresses the issue of business profit taxation. The paper states the following:

(12)

“The profits of an enterprise of a Contracting State shall be taxable only in that State unless the enterprise carries on business in the other Contracting State through a permanent establishment situated therein. If the enterprise carries on business as aforesaid, the profits of the enterprise may be taxed in the other State but only so much of them as is attributable to that permanent establishment.” (Model Tax Convention on Income and on Capital 2010)

Therefore, an enterprise from one state will not be double taxed if it carries business in another state through a permanent establishment. At the same time, the enterprise will be taxed only on the profits generated in that State. In order for that enterprise to evade taxation from its home country it has to change its permanent establishment status. Thus, the legal framework of the OECD area permits legally to avoid tax by changing the source of profit generation and legal establishment.

2.4

Tax Effect on FDI

To what extent does taxation have an impact upon FDI decisions? This question has been the subject of a vast number of studies over the past 30 years and the answers which are provided differ widely (OECD, Corporate Tax Incentives for Foregin Direct Investment 2001). Some studies have considered the effects of tax systems generally, some have examined specific taxes (notably corporate income tax), and some have concentrated upon special tax incentives. Theoretical studies suggest that taxation ought to be very important, since it influences both

production costs and the net profits that are available for distribution. However,

econometric studies, which seek to establish the relationship, if any, between changes in taxation and levels of FDI in a particular country, are mostly inconclusive, no doubt because there are so many other variables that might influence FDI flows (Easson 2004). Conforming to Pirnia and Jacques (2000) the little importance of taxes does not mean that they do not have any impact on inflows of FDI and it is not a coincidence that FDI in some tax havens grew more than fivefold between 1985 and 1994, to over $200 billion.

According to Easson (2004) taxes generally:

Play little part in the initial decision to invest abroad:

Since taxation affects the net profit available for distribution or reinvestment, this might suggest that high tax rates would induce a company to invest abroad in economies where ETR is low.

May play a more important role in location decisions:

Taxes in potential host countries do come into consideration once the decisions to invest abroad have been made (Easson 2004). Most econometric studies say that investors are mostly influenced in their decisions by market and political factors and that tax policy appears to have little effect on the location of FDI. However it is fairly accepted that tax considerations do influence FDI decisions “at the margin” (UNCTAD 1996)

Are more important for some types of investment than for others:

Distinction between market oriented and export oriented investments should be pointed very carefully (Easson 2004). Market oriented FDI would seem to be relatively little affected by considerations of taxation except, perhaps, where the host country tax is unusually

(13)

burdensome, because taxes that affect the cost of production will also be borne by domestic and other MNE competitors and will normally be passed on to consumers. By contrast, it is widely accepted that export oriented FDI is more sensitive to the host country tax burden (Easson 2004). This finding is not really surprising because export-oriented firms such as clothing manufacturers are operating in highly competitive markets with very slim margins (Pirnia and Jacques 2000).

Are growing in importance:

Most of the empirical studies done before 1990 found that taxation was a relatively minor consideration in most FDI decisions (Easson 2004). More recent studies, however, have tended to suggest the opposite. According to an OECD, Corporate Tax Incentives for Foregin Direct Investment 2001:

”the results of this recent work indicate that the location of real capital by manufacturing firms is sensitive to taxation and has become more so over time”

For this phenomenon there are a number of explanations. One of them is: as trade barriers between countries are eliminated, the remaining obstacles become more important (Easson 2004).

(14)

3

Theoretical Framework

This section presents the theoretical framework of the analysis. It begins with the OLI theory developed by John Dunning in 1981. Afterwards, it continues with previous studies in relation to the tax effect upon the distribution of investments worldwide and the results of these studies are briefly discussed.

3.1

The Eclectic Theory of International Production

Decisions by MNEs to carry out foreign investment are usually complex mostly because that process involves strategic decisions. The most prominent theory of FDI is “The Eclectic Theory” developed by John Dunning (1981). If an MNE wants to maximise its profits and prevent losses, FDI would be implemented if the Ownership, Location and Internalization (OLI) conditions are met (Dunning 1981).

Figure 6 OLI Paradigm

According to De Mooij and Ederveen (2003) first, there must be “Ownership” advantage for the multinational relative to ownership by local firms. This may have something to do with specific technological or organizational knowledge of the multinational, but could also relate to tax issues. Second, it must be attractive for the MNE to produce abroad because of some comparative “Location” advantage. Otherwise the multinational would have chosen to export, rather than invest. Finally it should be attractive to undertake activities within the multinational, rather than buying or leasing them from other firms. Taxes can affect all three OLI conditions, including the tax treatment for foreign firm, relative to domestically owned firms. The tax rate may also partially determine the attractiveness of location for undertaking investments (De Mooij and Ederveen 2003).

3.2

Previous Studies

3.2.1 David Hartman (1985)

In his paper “Tax Policy and Foreign Direct Investment” (1985), David Hartman discusses the competition between countries in regard to taxes for foreign investments. He begins his study by presenting legal reforms that give the chance to multinationals to benefit from tax differences between countries. Specifically he presents the legal reform of home countries not taxing the benefit earned abroad as it would result in double taxation. At the same time he

FDI

Ownership

Location

(15)

states that the tax is not the only determinant of investments. He clarifies his point of view by saying that firms will profitably invest abroad as long as the after-tax return abroad exceeds the after-tax return at home. At the same time the author makes a distinction between the foreign investments as issuance of new stock and reinvestment of retained earnings in the host country. He points out, that tax is crucial for the choice between these two options. It is explained that the reinvestment of retained earning pays less tax than the issuance of new stock and transfer of funds because for the second option, the stock, has to pay home country tax (which supposedly is higher). Following the same logic Hartman suggests that higher home tax would reduce the amount of distributed profits for the home country and therefore decrease the tax revenue. One thought worth indicating is the fact that firms with mature foreign operations are not sensitive to home country taxation because tax for mature firms is an avoidable fixed cost (Hartman 1985).

3.2.2 Michael Boskin and William Gale (1987)

In their study “New Results on the Effects of Tax Policy on the International Location of

Investment” (1987), Michael Boskin and William Gale analysed the research conducted by

David Hartman and brought more recent evidence and thought to the issue of tax competition in regard to investments. Like Hartman, the authors begin their study by providing the foundation of tax incentives for investment location, specifically they speak about the Accelerated Cost Recovery System, which was meant to increase U.S. capital formation and stem the flow of U.S. investment abroad. They discuss the fact that the location of capital is not primarily motivated by the level of corporate income tax, but, as well, access to markets political considerations, labour costs and expected economic conditions. At the same time, Boskin and Gale consider that, nevertheless, tax is considered to have some effect upon the location of investments. The authors state that the Foreign Direct Investment is not directly interpreted as Domestic Net Investment because it does not include the inflows of funds that are used to purchase real capital assets. Therefore FDI should be interpreted as net foreign investment (Boskin och Gale 1987).

In contrast to Hartman’s study, the research conducted by Boskin and Gale is more statistical. They present statistical figures from the Bureau of Economic Analysis regarding United States of America. In their study they found that FDI had risen by 2000% from 1950 to 1984. At the same time, this growth was oscillating along the period which is explained by varying phases of шге U.S. economy. The statistic also shows a dominance of investments financed by debt and equity over reinvestment of retained earnings.

Their model is very similar to the model of Hartman and the elasticities are as well very close. The results present a negative elasticity of 2.9 between the ratio of FDI and GNP to the relative tax term. This means that an increase in the tax rate of 1% would decrease the level of incoming Foreign Direct Investments by 2.9 %, if everything else is kept constant. During the analysis, Boskin and Gale have concluded that the reinvestment of retained earnings is more sensitive to tax changes than the FDI financed by transfers (Boskin och Gale 1987).

Boskin and Gale’s research confirm the conclusion stated by Hartman and present strong evidence for the negative relationship between tax and FDI inflows even though their model is sensitive to the period and the functional form of the regression (Boskin och Gale 1987).

(16)

3.2.3 Joel Slemrod (1990)

Joel Slemrod (1990) has conducted a study - “Tax Effects on Foreign Direct Investment in

the United States: Evidence from a Cross – Country Comparison” in regard to the

relationship between tax and FDI as well. He starts his work by pointing the drawbacks of the previous studies on the same matter. Slemrod sets forward that previous studies have used the average tax rate and attempted to use only the host country tax. He considers that using the effective marginal tax rate for both the home and host country would yield more objective results. Besides all mentioned above the author regards all previous empirical results close to the ones provided by Hartman’s work (Slemrod 1990).

Another issue that Slemrod is concerned about is the interpretation and collection of the data for FDI. The Bureau of Economic Analysis describes FDI as “the earnings retained by subsidiaries and branches of foreign parents and transfers of funds from foreign parents to the U.S. firm, including both debt and equity”. The author considers that FDI does not correspond directly to any measure of real investment because it excludes the investments financed by funds raised locally. At the same time, the data for the previous studies was collected using benchmark surveys for 1959, 1974 and 1980. For nonbenchmark years the data was constructed by manipulating the benchmark information. At the same time, in 1974 the minimum ownership criterion for FDI was decreased from 25% to 10%. Some of the previous studies have ignored these circumstances which are of crucial importance to the study. Ignoring these issues would lead to subjective economic modelling. Slemrod had replicated the models from previous studies with the adjustments for the issues stated above and has found that the marginal tax rate has significant negative effect on transfers of funds but not on retained earnings. In order to get significant results for retained earnings, the author included more nontax variables in the model. Controlling for more variables results in significant positive results for the elasticity of tax in respect to FDI (Slemrod 1990).

Furthermore, Slemrod analyses the effect that tax has upon inward FDI for U.S. He states that U.S. tax rate has an effect upon investments coming from countries that have territorial tax policies, not source tax policies. Therefore, investments coming from countries that have foreign tax exemptions are more sensitive to changes in host country tax rates than investments coming from countries which offer tax credits (Slemrod 1990). Slemrod considers that the home country tax is very relevant to the analysis of the sensitivity of host country taxation to FDI. He motivates this by stating that the after tax return to investment in the home country is affected by the home country tax and because the home country tax rate is reflected in the effective tax rate for countries that have tax credit policies (Slemrod 1990). Moreover, the author carries on empirical data analysis of the model presented above for different categories of investing countries. The results suggest weak relationship between tax and FDI and lack of difference between taxation on transfers for countries which benefit from tax exemptions and countries which benefit from tax credits. The author explains this similarity of the sensitivity for the two taxing systems by unreliability of the data and the ability of firms from tax credits countries to engage in financial transactions that would benefit to a higher extent than expected from the tax rate differences (Slemrod 1990).

As a conclusion, Slemrod points out that the host country tax rate has an ambiguous effect upon Foreign Direct Investment due to financial instruments that can be used in order to hedge the tax rate. In order to account for the relationship between tax and investment, more financial variables have to be taken into consideration (Slemrod 1990).

(17)

4

Empirical Analysis

4.1

Data and Methodology

The sample data consists of 25 OECD countries (see Appendix) for a period of 14 years from 1996 to 2009. Initially the sample was composed of 34 OECD member states for a period of 30 years from 1980 to 2009. The data was reduced mainly because of data unavailability. Specifically, the data for effective tax and FDI was not found for the period before 1990. At the same time, other determinants were lacking data for the period after 1990. In order to account for this scarcity, the length of the longitudinal data was shortened. Besides the problem of early data collection, some countries did not accumulate data for the parameters of the study before the year 2000. These countries were deleted from the sample because this would reduce the data by less than in the case of deleting periods from the sample. It has to be noted that Luxemburg was deleted from the sample because the data available for this country was missing until the year 2000 as Belgium and Luxemburg consisted of the same economic union and the data was collected for both countries together. For Belgium it was possible to find the complete data from other sources, therefore it was not removed. Besides Luxemburg, the United States of America was also removed because it presented outlying observations which skewed the statistical analysis.

In order to construct a model that would explain the variation on FDI and taxes it is necessary to control for variables that would influence the distribution of investments. In this study the determinants were selected in accordance to previous studies and availability of data. The variables that can influence both FDI inflows and ETR and that were used in this study are presented below. It is important to note that all the indicators are for host countries.

4.2

The Dependent Variable

The dependent variable (Y) is formulated as Net Foreign Direct Investment Inflow share of GDP rather than pure Net Foreign Direct Investment Inflow. Boskin and Gale (1987) and Slemrod (1990) are few of the researchers that use the investment ratio of GDP for their model. The ratio provides a very good interpretation of the model because an increase in FDI is relative to the increase in income.

Equation 2: Dependent Variable Formula

FDI

“Foreign direct investment are the net inflows of investment to acquire a lasting management interest (10 percent or more of voting stock) in an enterprise operating in an economy other than that of the investor.“ (OECD, OECD Benchmark Definition of Foreign Direct

Investment 2008)

It is the sum of equity capital, reinvestment of earnings, other long-term capital, and short-term capital as shown in the balance of payments. The series of flows shows net inflows (new investment inflows less disinvestment) in the reporting economy from foreign investors. Data are in current U.S. dollars.

(18)

GDP

GDP at purchaser's prices is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. Data are in current U.S. dollars. GDP is very important factor for location decisions because it captures the market size and at the same time contributes to the interpretation of the data in relative terms.

“A firm will have a high incentive to invest in a country where local demand is high. On the supply side, a bigger market should diminish the cost of firms due to economies of scale.”

(Azémar and Delios 2008)

4.3

The Determinants

Effective Tax Rate (ETR)

It represents the share levied on distribution of domestic source income to a resident individual shareholder, taking account of corporate income tax, personal income tax and any type of integration or relief to reduce double taxation. It is a more accurate measure because it considers the tax paid by companies as compared to the STR that does not consider the tax incentives undertaken by governments. At the same time, previous studies have relied on the ETR as a determinant of FDI, therefore, it is of importance to compare the results and elasticities between studies.

The selection of the tax for an econometric analysis is a complex issue. It is however observed that earlier studies use the STR and more recent studies use the Effective Marginal Tax or Average Effective Tax.

“In the previous literature, the disincentive to investment caused by the tax system is implicitly measured by an average tax rate, computed as total taxes paid dividend by a measure of profits. However, the incentive to undertake new investment depends on the effective marginal tax rate, which as is well known, can deviate substantially from an average tax rate concept.” (Slemrod 1990)

Steven Clark (2008) states in his paper “Assessing the FDI response to Tax Reforms and Tax Planning” a lot of work has to be done in order to determine what tax measures tend to be factored by investors when deciding to invest abroad.

Share of Internet Users (INET/POP)

Internet users are people with access to the worldwide network. It is adequate to use the share of the population that is using the internet rather than just internet users because the population of internet users is dependent on the overall population. It is therefore assumed that a large number of internet users represent a sign of developed infrastructure. Infrastructure is crucial to the investment location decision because excellent roads, bridges and production utensils can influence negatively transportation costs and labor costs and as a result increase the revenue of investing companies.

In other studies, the length and quality of roads are used as a proxy for infrastructure, given that the countries analyzed in this paper are developed and that previous studies were done in

(19)

periods before the Dotcom boom, it is more representative in our opinion to use internet users as a proxy.

Equation 3: Share of Internet Users Formula

Wadhwani (2000), a member of the Bank of England’s monetary policy committee, has stated that internet is expected to have a high influence upon the productivity, margins and inflation over the next years. Therefore, it would eventually add to the competitiveness in attracting FDI. At the same time, Changkyu Choi (2003) states:

The Internet can improve the productivity in several ways. First, the Internet can lower prices by lowering search costs.

Trade Openness (TO)

Trade openness represents the degree to which countries permit or have trade with other economies. This variable is crucial for investment locations because, it accounts for export-oriented investments. Therefore, if the host economy is more opened towards trade then investing economies will gain more because of higher levels of exports and therefore have higher incentives to invest. Trade openness is calculated using the following formula:

Equation 4: Trade Openness Formula

- Imports of goods and services (IMP)

Imports represent the value of all goods and other market services received from the rest of the world. They contain the value of merchandise, freight, insurance, travel, transport, royalties, license fees and other fees. They exclude compensations for employees and investment income. The data used is in current U.S. dollars.

Firms whose activities are export oriented or import inputs should be attracted to a country’s degree of openness to international trade (Azémar and Delios 2008).

- Exports of goods and services (EXP)

Exports represent all the transactions between residents of a country and the rest of the world involving a change of ownership from residents to non-residents of general merchandise, goods sent for processing and repair, nonmonetary gold and services. The data used is in current U.S. dollars.

Labor cost (LC) * Exchange Rate (EX_RATE)

Labor cost represents the cost of wages paid to workers during accounting period on daily, weekly, monthly, or job basis plus payroll and related taxes and benefits. The data used is the total labor cost. The labor cost is of great importance to investment location decisions because

(20)

cheaper labor cuts down the cost of production and therefore contributes to a higher income for the investing company. The data obtained is in national currency, in order to obtain objective results; it has to be transformed into U.S. dollars because all the other monetary parameters are in U.S. dollars.

Controlling for wages, it is generally hypothesized that lower labor costs should encourage “efficiency-seeking” FDI flows. If this relationship is not always established for industrial countries, results for developing countries indicate that wage costs are significant determinant of FDI location (Azémar and Delios 2008).

The Exchange Rate represents the rate at which one currency may be converted into another. The rate used in this study is the exchange rate of the national currency of the 25 sample countries and the U.S. dollar. The rates are consistent over time which means that a conversion was made from national currencies to the Euro for Euro zone economies. The exchange rate is very important for multinationals when deciding to invest abroad because it can influence the relative wage and production cost of the host country compared to the home country. This would eventually lead to an improved rate of return. When the exchange rate is taken into consideration it has to be understood that its variation can be the result of a changing cost of financing. The exchange rate has been used for computing the dollar value of the labor cost variable.

On the one hand, a host country’s weak currency can make firms more likely to invest in the country because the local acquisition costs will be lower. On the other hand, in developing countries, a weak currency can be seen as a signal of instability and generates risk aversion

(Azémar and Delios 2008).

Long Term Interest Rate (R)

Long term bonds are instruments whose yield is used to represent the interest rate. The aim of this data was to capture the rate of return on country level. Because no such data is collected and the data regarding the rate of return on the national stock markets is not available for all the sample countries. It has to be noted that the interest rate does not fully embody the return on capital because it does not take into consideration the risk of the investment. Therefore, in order to control for risk we used inflation as a determinant of economic stability. At the same time, the 10 year bond interest rate captures the effect that the cost on debt has upon the investments, which is of crucial importance to FDI financed by debt.

Inflation

Due to the fact that the data used for previous variables is calculated in current $US, there is a need to account for the change in purchasing power of the currency over time. At the same time, inflation explains the stability of an economy. Therefore if the inflation rate is high this is a sign of an instable economy. Inflation is not included in the model because it does not have a very strong explanatory power. It is also considered highly correlated to the interest rate which is already included in the model.

(21)

4.4

Relationship Between the Determinants

For the sake of understanding the sensitivity of investments to taxes it is very important to observe the relationship between the other determinants. Therefore the correlation matrix of the variables is presented below:

Table 2:Correlation Matrix

ETR FDI/GDP LC TO R INET/POP GDP INFL

ETR 1.00 FDI/GDP 0.10 1.00 Labor Cost 0.26 -0.12 1.00 Trade Openness -0.15 0.48 -0.43 1.00 Interest Rate -0.08 0.02 -0.15 -0.05 1.00 Internet/Population -0.11 -0.05 0.26 -0.15 -0.45 1.00 GDP 0.25 -0.13 1.00 -0.44 -0.14 0.26 1.00 Inflation -0.39 0.054 0.51 0.21 0.59 -0.23 -0.11 1.00

As it can be observed from the table, there is no strong correlation between the variables. A very important part of the correlation matrix is the correlation between the ratio of FDI and the other determinants. As it can be seen, the ratio of FDI has a relatively strong correlation with Trade Openness of 0.48. As well, inflation has a strong correlation with the interest rate which would be expected. Due to problems of multicolinearity, the inflation rate has not been added to the models. The correlation between effective tax and FDI ratio is 0.10. This contradicts the previously mentioned theories that high tax should discourage FDI inflows.

4.5

Functional Form

In order to understand the relationship between ETR and FDI it is necessary to find a reliable functional form. The functional form constructed is based on previous studies as well as small purposeful changes not covered in the literature but which contributes to the model from the sample data. The relationship between FDI and Tax varies at different levels of tax. At low levels of tax the data is more compact and at high values of tax the data has a high variance. It can also be observed that the data contains outliers. In order to fit the data it is necessary to transform it into logarithmic function.

Logarithmic Function

The model is constructed as a logarithmic function because it provides good interpretation, compresses large values and it controls for heteroscedasticity. First of all, the logarithmic model provides good interpretation of the coefficients. The interpretation of the coefficients is in the following way: one percentage change in tax contributes to beta (the value of β in the regression model) percentage change in the share of FDI to GDP. Secondly, the logarithmic function compresses large values and as a result outliers are less problematic.

(22)

As it can be observed from the histogram from the appendix Figure 9 Histogram of Effective Tax below the average ETR (30%) and Figure 10 Histogram of Effective Tax above the average ETR (30%), the probability distribution of the ETR is normally distributed for values higher than 30% but it is not normally distributed for values lower than 30%. It has to be noted that the FDI data contains negative values, in order to log it, it was necessary to add the constant minimum number of the data in order to have only positive values. The same modification was made in Boskin and Gale’s study (1987).

FDI/GDP

The dependent variable is formulated as Net Foreign Direct Investment Inflow share of GDP rather than pure Net Foreign Direct Investment Inflow. Boskin and Gale (1987) and Slemrod (1990) are one of the few researchers that use the investment ratio of GDP for their model. The ratio provides a very good interpretation of the model because an increase in FDI is relative to the increase in income. Finally, the functional form with this ratio as dependent variable provides lower p-value for the coefficients and higher adjusted R-squared. This can also be understood from the scatter plot (see Figure 4) presented below. The figure presents the pooled information from all 14 years for all the countries on the observations of logged ETR and logged FDI/GDP. Therefore in the figure are 350 observations, each observation corresponds to a certain country in a certain year.

Figure 7 Log(ETR) to Log(FDI/GDP)

Trend Variable (T)

The trend variable represents data that gives each year a one step increase in value compared to the last year and it starts with the value one for the year 1996. The trend variable has been added to the model because it accounts for technological change, improvement in health-care and so forth and because it increases the significance of the model.

Dummy Variable (DUM)

For the purpose of estimating a good model, it is necessary to add dummy variables for the slope and intercept of effective tax elasticity in relation with FDI to control for the spread

2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 -10 -8 -6 -4 -2 0 2 LOG_FDI_GDP L O G _ E F F _ T A X

(23)

relationship. As it was stated before, the tax rate has different distributions at levels upper or lower than 30%. In order to account for this, it is necessary to acknowledge a model that provides different slopes and intercept for the effect of tax on FDI/GDP.

Autoregressive Process (AR)

After conducting a residual analysis it has been revealed that the model suffers from first order serial correlation. It is presented in Figure 15 Model 1 Initial test for Autocorrelation and Figure 18 Model 2 Initial Test for Autocorrelation

that the models have very significant coefficients for the lagged residuals. In order to correct for this, it is necessary to fit an Autoregressive process into the model Figure 13 Model 1 Autoregressive and Figure 17 Model 2 Autoregressive

4.6

Regression Model and Results

In order to analyze a sample of 25 countries over a period of 14 years, the most appropriate method is Ordinary Least Squares estimates for pooled panel data because it contracts all the information into a model (unlike the case of cross-sections or time series where the model would have 14 or 25 different coefficients depending on the model uses) and because it is used in previous studies. At the same time, it is irrelevant to use Generalized Least Squares because it is a good method to use if the data exhibits heteroscedasticity, whereas for the Ordinary Least Squares model, the heteroscedasticity problem has been solved by the logarithmic function.

The central model used for the statistical analysis is the following: Model I

Equation 5: Formula for Model I

where FDI/GDP is the ratio of FDI inflow as a ratio of GDP; ETR is the effective tax rate;

TO is trade openness; LC*EX_RATE is the labour cost multiplied by the exchange rate; R is

the 10 year bond interest rate; INET/POP is the share of internet users from the population and T is the trend variable. Model I presents the best goodness of fit and coefficient significance. At the same time, in order to control for varying intercept and slope at different levels of effective tax, the following model was constructed:

Model II

Equation 6: Formula for Model II

Where DUM represents a dummy variable for tax levels, it has a value of 1 for tax rates higher than the mean 31, and 0 otherwise. The results of the regressions are presented below (Table 3 Regression Output).

(24)

Table 3: Regression Output

Variable Model I Model II Model I AR Model II AR

Constant Coefficient Std Dev P-value 5.448795 (0.416274) 0.0000 4.555836 (0.554518) 0.0000 3.702067 (0.543051) 0.0000 3.172088 (0.630189) 0.0000 Effective tax (ETR) Coefficient Std Dev P-value 0.418285 (0.112912) 0.0003 0.638487 (0.154456) 0.0000 0.340082 (0.116331) 0.0038 0.505388 (0.158717) 0.0016 Trade Openness (TO) Coefficient Std Dev P-value 0.502677 (0.057844) 0.0000 0.551410 (0.059568) 0.0000 0.382605 (0.065992) 0.0000 0.428583 (0.069446) 0.0000 Labor Cost * Exchange

rate (LC*EX_RATE) Coefficient Std Dev P-value -0.713056 (0.024325) 0.0000 -0.691700 (0.025195) 0.0000 -0.504477 (0.049100) 0.0000 -0.502557 (0.049037) 0.0000 Long Term Interest

Rate (R) Coefficient Std Dev P-value 0.235874 (0.069296) 0.0008 0.191441 (0.070179) 0.0068 0.165270 (0.072616) 0.0236 0.137871 (0.073542) 0.0619 Internet Users (INET/POP) Coefficient Std Dev P-value 0.125763 (0.036003) 0.0006 0.095131 (0.037530) 0.0118 0.074360 (0.040535) 0.0677 0.053589 (0.042057) 0.2037 Dummy Effective Tax

(DUM) Coefficient Std Dev P-value 3.647158 (1.301134) 0.0054 2.699358 (1.447122) 0.0632 Effective Tax*Dummy (ETR*DUM) Coefficient Std Dev P-value -1.052502 (0.369764) 0.0047 -0.781012 (0.410674) 0.0583 1st difference Lagged FDI/GDP Coefficient Std Dev P-value 0.278330 (0.059090) 0.0000 0.257777 (0.059691) 0.0000 1st order Lagged

Resi-dual Coefficient Std Dev P-value 0.306375 (0.060347) 0.0000 0.277188 (0.061144) 0.0000 0.177362 (0.170938) 0.3005 0.166778 (0.179968) 0.3550 Trend ( T) Coefficient Std Dev P-value -0.017458 (0.009268) 0.0606 -0.018737 (0.009233) 0.0433 -0.017493 (0.009630) 0.0704 -0.019279 (0.009718) 0.0483 Adjusted R-squared 0.888017 0.893456 0.890618 0.891507 F-statistic 398.8198 307.1293 327.8527 257.5580

(25)

Model I (Figure 12 Model 1): The regression starts at 5.45 level of FDI/GDP. Following the

model if everything is kept constant, if ETR increases by 1% then FDI/GDP increases by (on average) 0.41 %. This contradicts, the previously mentioned studies that state that lower tax encourages investments into host country.

In order to fully understand the effect tax has on FDI, it is necessary to understand how other variables influence investments. From the table it can be observed that Trade Openness, Long Term Interest Rates and the share of internet users has a positive effect on the level of FDI/GDP and Labor Cost has a negative effect upon the level of FDI/GDP. The relationship between the determinants and the ratio of FDI is as expected. It is crucial pointing out that the model explains 88% of the sample behavior.

As it has been revealed by the results of the OLS regression, the ETR has an overall positive effect upon the share of FDI. Therefore, previous studies do not support the empirical results. In order to be able to interpret and understand the statistical results it should be understood that some countries have been excluded from the analysis. For example, Luxemburg has been deleted from the sample because it was not found enough data to run the regression for this country. It is worth mentioning although, that Luxemburg is regarded as a “tax heaven” and would have had a significant impact upon the results. Besides the exclusion of countries from the sample, the sample period has been cut down.

Nowadays, the world is so globalized that many countries have adopted other incentives to attract capital besides tax. In order to fully cover other reasons to invest abroad it is necessary to do a more qualitative research.

Model II (Figure 16 Model 2): The regression starts at the level of 4.55 FDI/GDP if the ETR

is below the average of 31. If the tax rate is above, the intercept is 8.20. Tracking the model, if the tax rate is below the mean then 1% increase in ETR provokes a 0.63 % increase in the ratio of FDI, if everything is kept constant. If the ETR is above the average – then 1% increase in effective tax provokes decrease of 0.41% in the ratio of FDI, if everything is kept constant. Therefore, it can be understood from the model, that at low levels of tax, the ratio of FDI is positively sensitive to tax. At high levels ETR and FDI ratio have negative relationship. The other variables have kept the same relationship with the ratio of FDI, only at different intensities. This model explains 89 % of the behavior of the sample.

Model II reveals that the ETR has a different effect upon investments at different levels of tax. A reasonable explanation for this finding can be that at low levels of tax, multinationals from relatively low tax rate countries value the cost of changing their permanent establishment in order to avoid a tax increase not to be justified by the revenue the company can save by moving to a jurisdiction with an even lower tax. Alternatively, in other companies with permanent establishment in relatively high tax countries the value of changing their establishment to a lower tax country is more rewarding due to the higher level of shielded revenue from tax.

Previous studies state that corporations are willing to invest abroad as long as the cost of doing so is lower than the cost they would face at home. Therefore, in order to fully explain the variation in investment and taxes it is necessary to compare the host and home country tax levels. The host tax may have a positive relationship in relation with FDI but may have negative elasticity if it were subtracted from the home tax level.

(26)

It can be pointed out by now that the tax rate is crucial for investment location decisions. All previous studies have presented reasons to believe that FDI could be influenced by the tax level, at the same time, it was shown how certain taxes can be avoided by corporations. All of these factors have to be taken in consideration when interpreting the empirical results shown in this paper. The statistical research done in previous studies generally present a negative relationship between taxes and FDI or an insignificant relationship. The empirical analysis for OECD countries done in this paper has generally provided significant positive elasticity results between taxes and FDI, which contradicts the thought that countries with lower taxes would be expected to attract more capital because tax decreases the income of multinationals. Although, the models present very interesting and significant results, the results may be biased due to serial autocorrelation because in both models the residuals are correlated with their lagged value. As it is presented in the table above, the models present significant coefficients for the lagged value of the residuals. In order to correct for this it is assumed that the ratio of FDI is an Autoregressive Process (AR). Therefore, an AR model was fitted to the existing models. The residual regressions of the AR present insignificant coefficients for the lagged values of the residuals which mean that the serial autocorrelation of the residuals was corrected.

Model I Autoregressive (AR) (Figure 13 Model 1 Autoregressive): The regression starts at

3.70 level of FDI ratio. Everything being kept constant, if ETR increases by 1% then FDI/GDP increases by 0.34%. All the other variables have the same relationship with FDI/GDP but at lower intensities. This can be explained by the fact that the first order lagged dependent variable contributes to the model and takes a share of the elasticity explained in the previous model by the variables. This model explains 89 % of the sample behavior.

Model II Autoregressive (AR) (Figure 17 Model 2 Autoregressive): The intercept of the

model is 3.17 if the ETR is below the mean, if it is above, the intercept is 5.87. If all the va-riables are kept constant and the ETR is below the average, then if the ETR increases by 1% then the ratio of FDI increases by 0.5%. If the ETR is above the average, 1% increase in the ETR provokes a decrease of 0.28 % in FDI/GDP. For the other determinants, the relationship is the same to the ratio of FDI, only at lower elasticity levels. This can be explained the same way as for Autoregressive Model I. Model II AR explains 89% of the sample behavior.

(27)

5

Conclusions

The defiance between this study and previous work can be explained by the fact that this

study is done on developed countries, the investment determinants have shifted over the years and the home tax is not taken into consideration.

First, the sample does not contain data for developing countries. The majority of previous studies have used the developing world as a sample. Thus, the effect of taxation can vary between developed and developing countries. It is widely accepted that developing countries have higher incentives to lower their tax in order to attract more capital than developed countries because poor countries are more concerned with economic growth and developed countries care more about economic stability. At the same time, most of the tax heavens are developing countries, and if the data for these countries were included in the sample, the results would probably change significantly. Secondly, the priorities for the investment have changed over the years. Most of the studies done on the topic of taxation and investment have been conducted before 1990.

Secondly, multinationals may pay more attention to other determinants, which may be positively correlated with tax. Recent studies have found a positive relationship between taxation and economic growth. “Contrary to what would be expected, taxes on corporate

income as a share of total taxation have a positive, though fragile, correlation with economic growth” (Widmalm 2001). Nowadays, corporations may be more interested in investing in

economies that are expected to grow. Therefore, if higher taxes have a positive effect upon economic growth, then they have a positive effect upon inward investments, as well.

Finally, the home taxation is not taken into consideration. Previous studies state that corporations are willing to invest abroad as long as the cost of doing so is lower than the cost they would face at home. Therefore, in order to fully explain the variation in investment and taxes it is necessary to compare the host and home country tax levels. The host tax may have a positive relationship in relation with FDI but may have negative elasticity if it were subtracted from the home tax level.

Another issue that this study reveals is that the tax elasticity to FDI varies at different levels of tax. The ETR has a significant positive relationship with investments at levels of ETR

lower than 31% and a significant negative relationship at levels of ETR higher than an effective tax rate of 31%.

At the same time, as it is pointed out in Figure 20 Model Effective tax <30 % and Figure 21 Model Effective tax >30 %, taking into consideration the data that is associated with a tax level lower than 30, the models present significant results for tax. If, on the other hand, we consider the data for tax levels higher than 30 then we observe that tax is not significant as an investment location determinant.

5.1

Suggestions for Further Research

A good direction of the research could be done by examining the effect of tax policy on the different FDI components (Greenfield, reinvested earnings, and M&As). There have not been much studies on this topic, because most of the authors prefer to focus on the level of total FDI in the country. Another interesting study could be to analyze the varying tax elasticities and significance at different levels of tax towards FDI and the reasons behind that phenomenon.

(28)

Acknowledgements

We would like to thank Ph.D. Candidate Sofia Wixe and Prof. Lars Petersson for their valuable advices and support throughout the process of writing this paper.

References

Azémar, Céline, and Andrew Delios. "Tax competition and FDI: The special case of developing countries." Journal of the Japanese and International Economies, 2008: 85-108. Boskin, Michael J., and William G. Gale. "New Results on the Effects of Tax Policy on the International Location of Investment." In The Effects of Taxation on Captial Accumulation, by Martin Feldstein, 201-222. University of Chicago Press, 1987.

Choi, Changkyu. "Does the Internet stimulate inward foreign direct investment?" Journal of

Policy Making, 2003: 5.

Clark, Steven. Assesing the FDI response to Tax Reforms and Tax Planning. 2008.

De Mooij, Ruud A., and Sjef Ederveen. "Taxation and Foreign Direct Investment: A Synthesis of Empirical Research." International Tax and Public Finance, 2003: 673-693. Devereux, Michael P., Ben Lockwood, and Michela Redoano. "Do countries compete over corporate tax rates?" Journal of Public Economics 92, 2008: 1210–1235.

Dunning, John H. International Production and the Multinational Enterprice. London: Allen & Unwin, 1981.

Easson, Alex. Tax Incentives for Foreign Direct Investments. The Hague, London, New York: Kluwer Law International, 2004.

Hartman, David G. "Tax Policy and Foregn Direct Investment." Journal of Public Economics, 1985: 107-121.

KPMG. KPMG's Corporate and Indirect Tax Survey 2010. KPMG International Cooperative, 2010.

Model Tax Convention on Income and on Capital. Paris: OECD, 2010.

Moosa, A. Imad. Foreign Direct Investment: Theory, Evidence and Practice. London, 2002. OECD. Corporate Tax Incentives for Foregin Direct Investment. Paris: OECD, 2001.

OECD Factbook. Paris: OECD, 2010.

OECD. OECD Benchmark Definition of Foreign Direct Investment. OECD, 2008. OECD. OECD Factbook . Paris: OECD, 2010.

OECD. Recent Trends in Foreign Direct Investment. Paris: OECD, 2000.

Pirnia, Neda, and Morisset Jacques. How Tax Policy and Incentives Affect Foreign Direct

References

Related documents

If the immigration (Im) in a Swedish municipality increases with 1%, with all other regressors held constant, the Gini coefficient will on average increase with 0.1046% So, the

För det tredje har det påståtts, att den syftar till att göra kritik till »vetenskap», ett angrepp som förefaller helt motsägas av den fjärde invändningen,

The third analysis was done using Hofstedes (2005) model. According to this analysis Russia has vary numbers at all indexes. Russia has an extreme high level of

Human capital and Trade are not statistically significant they have a negative effect on economic growth.. Life expectancy has a positive effect on

As Regulatory Quality is significant also for the upper middle income group, it seems that the factors related to the business and investment climate have a higher threshold than

We expect trade openness to have a positive effect on economic growth given that the trade theory sees openness as beneficial and that previous research has found a

In regional gravimetric geoid determination, it has become customary to utilize the modified Stokes formula, which combines local terrestrial data with a global geopotential

The study finding shows that trade openness, gross domestic product, inflation, and lag of FDI are the most significant determinants of foreign direct investment inflows to