barrios et al. (2012) international taxation and multinational firm location decisions
TRANSCRIPT
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
1/13
International taxation and multinational rm location decisions☆
Salvador Barrios a, Harry Huizinga b,⁎, Luc Laeven c, Gaëtan Nicodème d
a European Commission, Joint Research Centre, IPTS b Tilburg University and CEPRc International Monetary Fund, Tilburg University and CEPRd European Commission, ULB, CEPR and CESifo
a b s t r a c ta r t i c l e i n f o
Article history:
Received 3 November 2008Received in revised form 18 June 2012
Accepted 18 June 2012
Available online 14 July 2012
JEL classi cation:
F23
G32
H25
R38
Keywords:
Corporate taxation
Withholding taxes
Dividends
Location decisions
Foreign direct investment
Multinationals
Using a large international rm-level data set, we examine the separate effects of host and additional parent
country taxation on the location decisions of multinational rms. Both types of taxation are estimated to have
a negative impact on the location of new foreign subsidiaries. The impact of parent country taxation is esti-
mated to be sizeable consistent with its international discriminatory nature. Our results show that interna-
tional double taxation by the parent country – despite the general possibility of deferral of taxation until
income repatriation – is instrumental in shaping the structure of multinational enterprise.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
With globalization and the progressive removal of barriers to trade,
an increasing number of companies develop international activities. To
access foreign markets, rms face a choice between producing goods at
home for exports and producing abroad. A host of tax and non-tax fac-
tors affect the decision whether to relocate production abroad. Among
the non-tax factors are the size of a foreign market, its growth pros-
pects, wage and productivity levels abroad, the foreign regulatory
and legal environment, and distance from the parent country (see
Görg and Greenaway (2004), Barrios et al. (2005), and Mayer and
Ottaviano (2007) for recent reviews). The impact of taxation on for-
eign direct investment (FDI) has been the subject of a sizeable liter-
ature, as reviewed by De Mooij and Ederveen (2006) and Devereux
and Maf ni (2007).
Studies of the effect of taxation on FDI location decisions generally
examine host country taxation to the exclusion of possible additional
parent country taxation. The contribution of this paper is to jointly con-
sider the impact of host and additional parent countrytaxation on mul-
tinational rm location decisions. As a rst level of taxation, the host
country may impose corporate income taxation on the income of localforeign subsidiaries. In addition, the host country could levy a non-
resident dividend withholding tax on the subsidiary's earnings at the
time they are repatriated to the parent rm. But taxation need not
stop at the host country level. The parent country can further choose
to levy a corporate income tax on the resident multinational's foreign
source income giving rise to internationaldouble taxation. We examine
the independent impactof all three levelsof taxation on the location de-
cisions of European multinationals over the period 1999–2003. Speci-
cally, we examine how these three levels of taxation affect in which
country a multinational headquartered in a certain country chooses to
locate a new foreign subsidiary.
Journal of Public Economics 96 (2012) 946–958
☆ The authorsthank JimHines andJoel Slemrod (the Editors), twoanonymous referees,
Wiji Arulampalam, Peter Finnigan, Andreas Hauer, Vanesa Hernandez Guerrero and
seminar participants at the Bancode España, the European Commission, the SolvayBrussels
School of Economics and Management, ZEW Mannheim, the CESifo area conference on
public sectoreconomics in 2009,and theCentrefor BusinessTaxationof OxfordUniversity
Summer Symposium in 2009 for valuable comments. The ndings, interpretations, and
conclusions expressed in this paper are entirely those of the authors. They should not be
attributed to the European Commission or the International Monetary Fund.
⁎ Corresponding author at: Department of Economics, Tilburg University, 5000 LE
Tilburg, Netherlands. Tel.: +31 13 4662623.
E-mail address: [email protected] (H. Huizinga).
0047-2727/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jpubeco.2012.06.004
Contents lists available at SciVerse ScienceDirect
Journal of Public Economics
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j p u b e
http://dx.doi.org/10.1016/j.jpubeco.2012.06.004http://dx.doi.org/10.1016/j.jpubeco.2012.06.004http://dx.doi.org/10.1016/j.jpubeco.2012.06.004mailto:[email protected]://dx.doi.org/10.1016/j.jpubeco.2012.06.004http://www.sciencedirect.com/science/journal/00472727http://www.sciencedirect.com/science/journal/00472727http://dx.doi.org/10.1016/j.jpubeco.2012.06.004mailto:[email protected]://dx.doi.org/10.1016/j.jpubeco.2012.06.004
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
2/13
We have collected new detailed information on how parent country
tax systems interact bilaterally with corporate taxation and non-
resident withholding taxation in the host country. Specically, we use
information on whether or not countries tax the income of their multi-
nationalson a worldwide basis, and whether foreigntax credits arepro-
vided for non-residentwithholding taxes only or also for theunderlying
host country corporate tax (as, for instance, in the United States). As an
alternative to worldwide taxes, parent countries may partially or fullyexempt foreign source income from taxation. For example, Germany
exempts 95% of the foreign source income of German multinationals
from taxation.
Data on the structures of European multinationals, and in particular
on newforeign subsidiarylocations, areobtained from theAmadeus da-
tabase. This data set allows us to consider multinational companies
resident in a broad set of countries, each potentially having foreign sub-
sidiaries in many other countries. Thus, unlike earlier work this paper
considers multinational rm location choices in a multilateral setting
of N by N countries. In addition to being an innovative approach, this
multi-country framework is necessary to obtain suf cient variation in
additional parent country corporate taxation (not highly correlated
with host country corporate taxation) to be able to separately estimate
its impact on international location decisions.
Our results suggest that host country and additional parent coun-
try corporate income taxation both discourage the location of foreign
subsidiaries in a particular country. In our benchmark estimation, the
estimated negative impact of the two types of taxation – as derived
from statutory tax information – is about of equal size. Our nding
of a signicant role for the additional parent country tax in foreign
subsidiary taxation is new, and perhaps surprising. Many countries
allow parent country taxes on foreign source income to be deferreduntil dividend repatriation, reducing the scope for these taxes to af-
fect location decisions. All the same, we nd that foreign subsidiary
location – and the resulting international ownership pattern of sub-
sidiaries – is sensitive to additional parent country taxation. Multina-
tional rms act on an apparently signicant incentive to bring about
an international ownership pattern of subsidiaries that is internation-
ally tax-ef cient.
We perform several robustness checks to better understand the role
of host and parent country taxation in determining foreign subsidiary lo-
cation. We nd that location decisions of highly protable foreign sub-
sidiaries are less responsive to both host and parent country taxation
than location decisions of less protable foreign subsidiaries, perhaps be-
cause high protability reects location- or owner-specic rents that
would be destroyed by an alternative locationor nationalowner. Further,
Table 1
Corporate taxation and double tax relief methods for dividends received in European countries in 2003.
Corporate tax rate including local
taxes and surcharges (%)
Treatment of foreign dividends
from treaty countries
Treatment of foreign dividends
from non-treaty countries
Austria 24 Exemption Exemption
Belgium 33.99 Exemption (up to 95%) Exemption (up to 95%)
Bulgaria 23.5 Indirect credit Direct credit
Croatia 20 Exemption Exemption
Cyprus 15 Exemption Exemption
Czech Republic 31 Indirect credit DeductionDenmark 30 Exemption Exemption
Estonia 26 Indirect credit Indirect credit
Finland 29 Exemption Direct credit
France 35.43 Exemption (up to 95%) Exemption (up to 95%)
Germany 39.59 Exemption (up to 95%) Exemption (up to 95%)
Greece 35 Indirect credit Indirect credit
Hungary 19.64 Exemption Exemption
Iceland 18 Exemption Exemption
Ireland 12.5 Indirect credit Indirect credit
Italy 38.25 Exemption (up to 60%) Exemption (up to 60%)
Latvia 19 Exemption Exemption
Lithuania 15 Exemption Exemption
Luxembourg 30.38 Exemption Exemption
Malta 35 Indirect credit Indirect credit
Netherlands 34.5 Exemption Exemption
Norway 28 Indirect credit Indirect credit
Poland 27 Indirect credit Direct CreditPortugal 33 Direct credit Direct credit
Romania 25 Indirect credit Indirect credit
Russia 24 Direct credit No relief
Slovak Republic 25 Indirect credit No relief
Slovenia 25 Exemption Exemption
Spain 35 Exemption Indirect credit
Sweden 28 Exemption Exemption
Switzerland 21.74 Exemption Exemption
Turkey 33 Indirect credit Direct credit
United Kingdom 30 Indirect credit Indirect credit
Notes: Corporate tax rate denotes the statutory corporate tax rate including local taxes and surcharges. Statutory corporate tax rate in Estonia is 0% on
retained earnings but a distribution tax of 26% is applied on distributed pro t. Corporate tax rate of France includes a 3% social surcharge and a special
3.3% surcharge for large companies. Corporate tax rate of Germany includes a solidarity surcharge of 5.5%, an average deductible trade tax of 16.14%, and
an exceptional surcharge of 1.5%. Corporate tax rate of Hungary includes a deductible local business tax. Corporate tax rate of Ireland applies to trading
activities. For non-trading activities, the rate is 25%. Corporate tax rate of Luxembourg includes employment surcharges and local taxes. Corporate tax
rate of Switzerland applies to the canton of Zurich and includes cantonal and local taxes in Zurich. Treatment of foreign dividends refers to double tax
relief convention used by parent country. Foreign subsidiaries are assumed to be fully owned. The indirect credit system applies to foreign dividendsfrom treaty countries in Poland as long as the holding stake is at least 75% for the past 2 years and there exists a bilateral treaty or the EU
parent-subsidiary directive applies. In Portugal, foreign dividends from treaty countries are exempt from corporate taxes if the EU parent-subsidiary di-
rective applies, but the foreign withholding tax is not creditable. In some countries, dividend income is exempt from taxes up to a certain percentage,
indicated between brackets. Source: International Bureau of Fiscal Documentation.
947S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
3/13
location choicesof foreign subsidiarieswithlowxed assetsare more re-
sponsiveto both host andparent country taxation. This could reect that
high xed assets are a barrier to choosing an alternative physical loca-
tion. Finally, we nd that international location decisions are relatively
sensitive to international double taxation – including the additional par-
ent country taxation– if theparent country does notallowthe deferralof
foreign source income until dividend repatriation.
Existing studies of the inuence of corporate taxes on multina-
tionals’ location have in general paid little attention to the role played
by parent country taxes. For instance, Devereux and Grif th (1998) in-
vestigate how host country taxationaffects the subsidiary location deci-
sions of US multinationals in several large European countries (France,
Germany, and the United Kingdom) over the period 1980–1994. Theynd that – conditional on the choice to locate production abroad –
host country average effective tax rates (but not marginal effective tax
rates) are important in determining foreign location choice, even if tax-
ation does not appear to affect the earlier choice to locate abroad or to
export. Buettner and Ruf (2007) in turn nd that location choices of
German multinationals across 18 potential host countries between
1996 and 2003 are affected more by host country statutory tax rates
thaneffective average tax rates, while they nd no effect of marginal ef-
fective tax rates.
Several authors, however, have previously found a role for addi-
tional parent country taxation to affect the location of FDI. For US
multinationals, Kemsley (1998) nds that the host country tax only
affects the ratio of US exports to foreign production over the period
1984–
1992 if the multinationals nd themselves in excess credit
positions.1 Analogously, a role of parent country taxation in affecting
FDI into the United States is found by Hines (1996) who shows that
foreign countries with worldwide taxation invest relatively much in
US states with high state taxes. This reects that multinationals locat-
ed in countrieswith worldwide taxation may be able to obtainforeign
tax credits for US state corporate income taxes. Altshuler and Grubert
(2001) evaluate the implications of a hypothetical switch to dividend
exemption by the US to conclude, in contrast, that there is no evi-
dence that such a change would materially affect international loca-
tion decisions of US multinationals. Egger et al. (2009) construct an
effective tax rate on a bilateral basis that reects overall host and
home country taxation, and nd that this bilateral effective tax rate
has a negative impact on bilateral FDI stocks after controlling forhost and parent country unilateral effective tax rates.2 However,
none of these papers studies a rm's location decision in a multilater-
al setting of N by N countries, as we do.
Table 2
Bilateral dividend withholding tax rates in Europe in 2003.
Sub sidiary c ountr y Par ent c ou ntr y A ustr ia Belgium B ulga ria C roat ia C yp rus C zech R ep . Denmar k Estonia Finla nd France Germa ny
Austria X 0 0 0 10 10 0 5 0 0 0
Belgium 0 X 10 10 10 5 0 25 0 0 0
Bulgaria 0 10 X 5 5 10 5 15 10 5 15
Croatia 0 10 5 X 10 5 5 15 5 5 0
Cyprus 0 0 0 0 X 0 0 0 0 0 0
Czech Rep. 10 5 10 5 10 X 15 5 5 10 5
Denmark 0 0 5 5 10 15 X 5 0 0 0Estonia 0 0 0 0 0 0 0 X 0 0 0
Finland 0 0 10 5 29 0 0 0 X 0 0
France 0 0 5 5 10 10 0 5 0 X 0
Germany 0 0 15 15 10 5 0 5 0 0 X
Greece 0 0 0 0 0 0 0 0 0 0 0
Hungary 10 10 10 10 5 5 5 20 5 5 5
Iceland 15 15 15 15 15 5 0 5 0 5 5
Ireland 0 0 0 20 0 0 0 0 0 0 0
Italy 0 0 10 10 15 15 0 5 0 0 0
Latvia 10 10 10 5 10 5 5 5 5 5 5
Lithuania 15 15 15 5 15 5 5 0 5 5 5
Luxembourg 0 0 0 20 20 0 0 20 0 0 0
Malta 0 0 0 0 0 0 0 0 0 0 0
Netherlands 0 0 5 0 25 0 0 5 0 0 0
Norway 5 15 15 15 0 5 0 5 0 0 0
Poland 10 10 10 5 10 5 5 5 5 5 5
Portugal 0 0 15 30 30 15 0 30 0 0 0Romania 10 5 10 5 10 10 10 10 5 10 10
Russia 5 15 15 5 0 10 10 15 5 5 5
Slovak Rep. 10 5 10 5 10 5 15 15 5 10 5
Slovenia 5 5 15 15 10 5 5 15 5 5 15
Spain 0 0 5 15 15 5 0 15 0 0 0
Sweden 0 0 0 0 0 0 0 0 0 0 0
Switzerland 5 10 5 35 35 5 0 35 5 5 5
Turkey 16.5 16.5 16.5 11 16.5 16.5 16.5 16.5 16.5 16.5 16.5
United Kingdom 0 0 0 0 0 0 0 0 0 0 0
Notes: Withholding tax rates apply to dividends paid by fully owned subsidiaries in subsidiary country to parent rm. Bilateral tax treaties are taken into account. Parent-Subsidiary
Directive is binding between EU Member States and provides exemption from withholding tax if equity holding is at least 25%. The reported gures assume an equity holding in the
subsidiary of at least 25%. In Ireland, subsidiaries owned by parent companies resident in EU or treaty countries are exempt from withholding tax provided that they are not under
the control of persons not resident in such countries. In Italy, authorities can provide a refund equal to the tax claimed limited to 4/9 of the Italian withholding tax if the recipient
can prove a tax is paid in his country on the dividend. In Luxembourg there is an exemption from withholding tax for EU and treaty partners if holding in company resident in
Luxembourg is at least 10%. Source: International Bureau of Fiscal Documentation.
1 US multinationals are subject to worldwide taxation in the United States. Thus,
they have to pay tax in the United States on their foreign-source income, subject to
the provision of a foreign tax credit for taxes already paid in the host country. The for-
eign tax credit, in practice, is limited to the amount of US tax due on the foreign-source
income. This implies that the overall tax on the foreign income is the host country tax if
this tax exceeds the US tax, while it is the US tax if this tax is the higher of the two. US
taxes on foreign source income can be deferred until the income is repatriated.2 Repatriation taxes more broadly can affect multinational rms' behavior. Desai et
al. (2001) analyze the effect of repatriation taxes on dividend payments by the foreign
af liates of US multinational rms to nd that 1% lower repatriation tax rates are asso-
ciated with 1% higher dividends.
948 S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
4/13
Using information on international M&As, Huizinga and Voget
(2009) nd that the international ownership pattern that emerges re-
ects an effort to avoid international double taxation of foreignsource income. Desai and Hines (2002) argue that international dou-
ble taxation imposed by the US can explain inversions of existing US
multinationals, whereby the original US parent becomes a subsidiary
and the earlier foreign subsidiary becomes the new parent rm. Voget
(2011) nds that a multinational is more likely to relocate its head-
quarters abroad, if it is located in a parent country with high taxation
of foreign source income. This varied evidence on how international
double taxation affects the structure of the multinational rm is
consistent with a pattern of foreign subsidiary location that avoids
additional parent country taxation, as found in the present paper.
Huizinga et al. (2012) demonstrate that the international double
taxes that are triggered by an international M&A are to a large extent
capitalized into a lower takeover price. This also suggests that inter-
national double taxes are economically burdensome to multinationalrms, providing them with an incentive to own foreign subsidiaries
that are subject to low parent country taxation.
At a theoretical level, Hartman (1985) makes a distinction between
‘mature’ foreign af liates that are nanced at the margin by retained
earnings, and ‘immature’ foreign af liates that rely on funding from
their parent rms. The af liate's required rate of return is shown to re-
ect parent country taxation only if it is immature. Parent country tax-
ation thus should be particularly important for the newly established
foreign subsidiaries that are examined in this paper.
In the remainder of this paper, Section 2 describes the tax treat-
ment of the foreign source income of multinational rms. Section 3
discusses our rm-level data. Section 4 presents estimates of the im-
pact of international taxation on the location of foreign subsidiaries.
Finally, Section 5 concludes.
2. The international tax system
This section describes the corporate tax system applicable to amultinational company with foreign subsidiaries.3 Consider a multi-
national company with a parent located in home country p and a sub-
sidiary located in host country s. Both home and host countries may
tax the subsidiary's income. First, the host country may levy a corpo-
rate income tax at a rate t s on this income. Table 1 shows the statutory
corporate income tax rates for the 33 European countries in our sam-
ple for the year 2003.4 These statutory tax rates are those on distrib-
uted prots and include local taxes and applicable surcharges. In our
sample, the corporate tax rate for 2003 ranges from a low of 12.5% in
Ireland to a high of 39.6% in Germany.
Next, the host country levies a non-resident dividend withholding
tax at a rate ws on the subsidiary's net of corporate tax income upon
repatriation of this income to the parent. Table 2 provides informa-
tion on the applicable withholding tax rates on dividends paid byfully owned subsidiaries to their non-resident parents in 2003. For
example, a dividend paid by a Belgian subsidiary to its parent compa-
ny located in Estonia will be subject to a withholding tax of 25%, while
the withholding tax on a dividend paid by an Estonian subsidiary to
its Belgian parent company has a zero rate. The withholding tax
Greece Hungary Iceland Ireland Italy Latvia Lithuania Luxembourg Malta Netherlands Norway Poland
0 10 25 0 0 25 25 0 15 0 5 10
0 10 25 0 0 25 25 0 15 0 5 10
10 10 15 5 10 15 15 5 0 5 15 10
5 5 15 15 10 5 15 15 5 5 15 5
0 0 0 0 0 0 0 0 0 0 0 0
15 5 5 5 15 5 5 5 5 0 5 5
0 5 0 0 0 5 5 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 0
0 5 0 0 0 0 0 0 0 0 0 0
0 5 5 0 0 5 5 0 5 0 0 5
0 5 5 0 0 5 5 0 5 0 0 5
X 0 0 0 0 0 0 0 0 0 0 0
10 X 20 5 10 20 20 5 5 5 10 10
15 15 X 15 15 5 5 5 15 0 0 5
20 0 20 X 0 0 0 0 20 0 0 0
0 10 27 0 0 27 5 0 15 0 15 10
10 10 5 5 10 X 0 10 5 5 5 5
15 15 5 5 5 0 X 15 15 5 5 5
0 0 0 0 0 20 20 X 0 0 0 0
0 0 0 0 0 0 0 0 X 0 0 0
0 5 0 0 0 5 5 0 5 X 0 0
20 10 0 0 15 5 5 5 15 0 X 5
15 10 5 0 10 5 5 5 5 0 5 X
0 15 15 0 0 30 30 0 15 0 15 1510 5 10 3 10 10 10 5 5 5 10 5
15 10 15 10 5 15 15 10 15 5 10 10
15 5 15 0 15 10 10 5 5 0 5 5
15 10 15 5 10 5 5 5 15 5 15 5
0 5 5 0 0 15 15 0 15 0 10 5
0 0 0 0 0 0 0 0 0 0 0 0
5 10 5 10 15 5 5 0 35 0 5 5
16.5 11 16.5 16.5 16.5 16.5 11 16.5 16.5 16.5 16.5 11
0 0 0 0 0 0 0 0 0 0 0 0
3 See Huizinga et al. (2008) for a more detailed description of corporate tax systems
as they apply to multinational companies. This study is limited to the impact of the in-
ternational taxation of dividends on location decisions, even though the taxation of
other forms of income such as royalties or interest could also be important in corporate
location choices.4 For illustrative purposes, the tables report taxation data for the year 2003 only, al-
though we have collected these data for the entire period 1999–
2003.
949S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
5/13
rates for transactions involving two EU Member States currently are
zero on account of the EU Parent-Subsidiary Directive.5
The net of withholding tax dividend received by the parent com-
pany is in principle taxed in the parent country–subject to some
form of double tax relief as recommended by the OECD Model TaxTreaty or as prescribed by the EU Parent-Subsidiary Directive. Some
countries operate an exemption system. In this instance, the dividend
is not taxed in the parent country, if the provided exemption is full.
The overall international rate of taxation on the subsidiary's income
is then given by 1−(1− t s)(1−ws) or t s+ ws− t sws.
In other instances, the home country may tax the worldwide in-
come of its multinationals and subject the received dividend to corpo-
rate income taxation at a rate t p. Generally, a foreign tax credit is
provided for taxes paid in the host country, usually limited to the
amount of the home tax due on the foreign source income. Some
countries apply an indirect tax credit system under which both the
corporate tax and the withholding tax paid in the host country are
credited against the home corporate income tax. In case the home
country's corporate income tax t p is higher than the overall hostcountry tax rate t s+ ws−t sws, the rm pays income tax in the
home country at a rate t p− [t s+ws− t sws] so that the combined, ef-
fective tax rate is equal to t p. If instead the home country's corporate
income tax rate is lower than the overall host country's rate, the rm
is said to be in excess foreign tax credit and it will pay no further tax
in the home country (having reduced its home tax liability to zero by
using foreign tax credits). In this instance, the combined, effective tax
rate is t s+ws− t sws. In summary, for home countries with an indirect
tax credit system, the combined, effective tax rate is equal to max
[t p, t s+ ws− t sws].
Home countries may restrict the foreign tax credit to cover only
host country non-resident withholding taxes giving rise to a direct
tax credit system. In this case, the multinational has to pay tax in
the home country to the extent that t p exceeds ws and the combined,
effective tax rate is given by t s+(1−
t s) max[t p, ws].Alternatively, some home countries offer neither exemption nor a
foreign tax credit for taxes paid abroad, but instead allow foreign
taxes to be deducted from home country taxable corporate income.
This amounts to the deduction system with a combined, effective
tax rate of 1−(1− t s)(1−ws)(1−t p).
Finally, in some rather exceptional cases no double tax relief is
provided at all. With full double taxation, the combined, effective
tax rate becomes t s+ ws− t sws+t p.
Columns 3 and 4 of Table 1 indicate which doubletax relief system
is applied by European countries in the sample. As seen in the table,
some countries provide different double tax relief to treaty partners
and non-treaty countries. Thus, we need to know whether there
exist doubletax treaties among thecountries in oursample. On a bilat-
eral basis, this information is provided in Table 3 with the value 1 in-dicating the existence of such a treaty and 0 its absence. The table
indicates that for many countries the treaty network is not complete.
Forexample,in 2003 theCzechRepublic hasa treaty with allcountries
in the sample except Malta and Turkey. From Table 1, we see that this
implies that dividends from all foreign subsidiaries paid to a Czech
parent benet from an indirect tax credit, except for those paid by a
Maltese or a Turkish subsidiary where the deduction system applies.
Information from Tables 1–3 allows us to calculate the combined
effective tax rate on foreign dividends for any pair of home and host
countries. To x ideas, consider the case of a dividend paid by a Mal-
tese subsidiary to its Czech parent in 2003. Table 1 shows that the
statutory corporate tax rate in Malta is 35%. We infer from Table 2
that net prots paid as a dividend to a foreign company are never
subject to a non-resident dividend withholding tax in Malta. As
Subsidiary country Parent country Portugal Romania Russia Slovak Rep. Slovenia Spain Sweden Switzerland Turkey United Kingdom
Austria 0 15 5 10 5 0 0 0 25 0
Belgium 0 5 10 5 5 0 0 10 15 0
Bulgaria 10 10 15 10 15 5 10 5 10 10
Croatia 15 5 5 5 15 5 15 5 10 5
Cyprus 0 0 0 0 0 0 0 0 0 0
Czech Rep. 10 10 10 5 5 5 0 5 15 5
Denmark 0 10 10 15 5 0 0 0 15 0
Estonia 0 0 0 0 0 0 0 0 0 0
Finland 0 0 0 0 5 0 0 0 15 0
France 0 10 10 10 5 0 0 5 15 0
Germany 0 10 5 5 15 0 0 0 15 0
Greece 0 0 0 0 0 0 0 0 0 0
Hungary 10 5 10 5 10 5 5 10 10 5
Iceland 10 15 15 15 15 5 0 5 15 5
Ireland 0 0 0 0 0 0 0 0 20 0
Italy 0 10 5 15 10 0 0 15 15 0
Latvia 10 10 10 10 5 10 5 5 10 5
Lithuania 15 10 15 10 5 5 5 5 10 5
Luxembourg 0 0 0 0 0 0 0 0 20 0
Malta 0 0 0 0 0 0 0 0 0 0
Netherlands 0 0 5 0 5 0 0 0 5 0
Norway 10 10 10 5 15 10 0 5 20 5
Poland 10 5 10 5 5 5 5 5 10 5
Portugal X 15 15 30 30 0 0 15 30 0
Romania 10 X 10 10 5 10 10 10 10 10
Russia 10 15 X 10 10 5 5 5 10 10Slovak Rep. 15 10 10 X 5 5 0 5 5 5
Slovenia 15 15 10 5 X 5 5 5 15 5
Spain 0 5 5 5 5 X 0 10 15 0
Sweden 0 0 0 0 0 0 X 0 0 0
Switzerland 10 10 5 5 15 10 0 X 35 5
Turkey 16.5 16.5 11 5.5 16.5 16.5 16.5 16.5 X 16.5
United Kingdom 0 0 0 0 0 0 0 0 0 X
Table 2 (continued)
5 Note that in 2003 prior to their accession, many new EU Member States still
maintained non-zero rates vis-à-vis EU countries and vice versa.
950 S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
http://-/?-http://-/?-http://-/?-http://-/?-
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
6/13
already mentioned, Table 3 indicates that no tax treaty was in force
on income repatriated from Malta to the Czech Republic in 2003 so
that from Table 1 we see that incoming foreign dividends benet
from a deduction system in Czech Republic. Finally, the same table
indicates that the applicable corporate tax rate in this country is
31%. From the formula above, the combined, effective tax rate equals
1−(1−0.35)×(1−0)×(1−0.31)=55.2%. This rate is considerably
higher than the Maltese corporate tax rate of 35%. This suggests
that the additional taxation of multinational
rms, in the form of withholding taxes and home country corporate income taxation, po-
tentially has an independent and signicant impact on international
location decisions.
Below, we will investigate the independent inuences of host
country corporate income and dividend withholding taxation and
home country corporate income taxation on corporate location deci-
sions. To make parameter estimates comparable across tax measures,
it is useful to construct all three tax measures as shares of the foreign
subsidiary's pre-tax income. The host country tax rate is already de-
ned as a share of the subsidiary's pre-tax income. Our withholding
tax measure will be (1− t s) ws to reect that the withholding tax ap-
plies to the subsidiary's income net of the host country corporate tax.
Finally, the additional parent country corporate income tax – as a
share of the subsidiary's pre-tax income – is computed as the differ-
ence between the combined, effective tax rate and t s+ ws− t sws.
We will dene the international tax to be the sum of the withholding
tax and additional parent country corporate tax both expressed as
shares of the subsidiary's pre-tax income. Equivalently, the interna-
tional tax is the difference between the combined, effective tax and
the host country corporate income tax.
Unlike host country corporateincome taxes, withholding taxesand
home country corporate income taxes are generally deferred until the
foreign source income is repatriated to the parent in the form of divi-
dends. Deferral reduces the present value of taxation. Thus, withhold-
ing taxes and home country corporate income taxes are expected to‘bite’ less than host country corporate income taxes. Whether the de-
ferral of withholding taxes and home country corporate income taxes
serves to make these taxes immaterial for location decisions is an em-
pirical matter. This is what we turn to in the empirical section below.
3. Multinational enterprise data
Data on the structure of multinational rms in Europe are taken
from the Amadeus database. This database provides standard ac-
counting data and data on ownership relationships within corporate
groups.6 We have data on multinational rms operating in 33 Europe-
an countries over the years 1999–2003. The ownership data enable us
to match European rms with their domestic subsidiaries and foreign
subsidiaries located in other European countries.7 We have owner-
ship information for the years 1999, 2001, and 2003. A rm is called
a subsidiary if at least 50% of the shares are owned by a single other
rm. A subsidiary is taken to be new in its year of incorporation.
In our benchmark analysis below, we consider 909 new foreign sub-
sidiaries. Information on the number of parent and subsidiary countries
involved in these new locations is provided in Panel A of Table 4. Our
benchmark sample excludes new locations where the parent company
becomes an intermediate company as it is a subsidiary itself of another
parent company. The United Kingdom with 115 new parent companies
has most new parent companies, followed by France with 84 new par-
ent companies. Eachsubsidiary has a home country (where its parent is
located) and a host country (whereit is located itself). For each country,the table lists the number of subsidiaries by home country and by host
country. The table indicates that, for example, France, the Netherlands,
and the United Kingdom are the home country to relatively many
subsidiaries. Hence,there are relatively manysubsidiaries with a parent
rm in one of these countries. Denmark, Germany and the United
Kingdom, on the other hand, are the host country to relatively many
subsidiaries.
Our subsequent empirical work on foreign subsidiary location
aims to predict the location of a new foreign subsidiary in 1 of 32 for-
eign European countries. The dependent variable, called subsidiary
location, takes on a value of one if a particular country is selected as
a subsidiary's location and it is zero otherwise.
Summary statistics on the subsidiary location variable, the tax vari-
ables, and some controls are provided in Panel B of Table 4 (see Table
A1 in the Appendix for variable denitions and data sources). The
26,648 observations reported in the table are identical to the number of
observations in the basic regression 1 of Table 5.8 The mean value of the
overall effective tax is 0.35. This mean effective tax, in effect, is the sum
of a mean host country tax of 0.30 and a mean international tax of 0.05.
Among the control variables, GDP bilateralis the ratio ofthe GDP of a
potential host country andthe sumof theGDPs of allother potential for-
eign (but not domestic) locations. This variable captures market size,
and it is expected to exert a positive impact on the probability of subsid-
iary location in a host country.
Contiguity is a dummy variable signaling a common border be-
tween host and home countries. A common border is expected to
make location in the host country more likely.
Difference in labor costs is the log of the ratio of labor costs in the
home country and labor costs in the host country, expressed in a com-mon currency. The impact of higher labor costs in the host country on
the probability of location is in principle ambiguous (see Kimino et al.
(2007) for a review of the empirical evidence). Higher labor costs in
the host country (or a lower difference in labor costs variable) are
expected to discourage location for a given level of labor productivity
in the host country. In contrast, they potentially encourage location to
the extent that they signal high labor skills and productivity that are
sought after by the multinational rm.
Economic freedom is an index of the extent of soundness of the legal
system, absence of trade barriers, absence of price controls, and transfers
and subsidies as a share of GDP. Economic freedom should make a coun-
try attractive as a subsidiary location.
Finally, EU membership is a dummy variable agging EU member-
ship of a prospective host country. EU membership, to the extent thatis signals commitment to high EU standards of dealing with foreign in-
vestors, could engender subsidiary location. Moreover, establishing a
subsidiary in one EU country allows non-EU rms to trade freely across
all other EU countries by taking advantage of the EU common market.
Panel C of Table 4 provides correlation coef cients among the lo-
cation, tax and control variables. Interestingly, location is positively
and signicantly related to the host country tax, but negatively and
signicantly to the international tax. The rst correlation possibly re-
ects that subsidiaries tend to be located in larger countries, which
tend to have relatively high corporate income taxes. In the table,
the host country tax is indeed positively and signicantly correlated
6 The database is created by collecting standardized data received from 50 vendors
across Europe. The local source for these data is generally the of ce of the Registrar
of Companies.7 The Amadeus database only contains information on European rms and we there-
fore only cover the European operations of the multinationals in our sample. In 2001,
intra-EU25 FDI ows amounted to 54% of all outward FDI ows from EU25 countries,
while intra-EU25 FDI stocks were 57% of outward FDI stocks of EU25 countries
(according to Eurostat). Thus, European multinationals have most of their FDI out-
standing in other European countries. All the same, our conditioning on a European lo-
cation of new foreign subsidiaries potentially affects estimated sensitivities of location
to taxation. In particular, by conditioning on European locations rather than locations
worldwide, we increase the expected probability of location in any one country and
potentially also the sensitivity of location to taxation. New foreign subsidiaries tend
to be large relative to the overall assets and capital of the multinational. In our sample,
a new foreign subsidiary on average represents 23% of the assets and 24% of the capital
of the expanded multinational rm including the new foreign subsidiary.
8 Note that the mean value of the location variable is not exactly 1/32 due to the ab-
sence of data for some specic combinations of countries and years.
951S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
7/13
with the GDP bilateral variable as an index of host country relative
size. The negative correlation between location and the internationaltax could reect that subsidiary location is chosen so as to mitigate in-
ternational double taxation. The host country tax and the internation-
al tax are negatively correlated, perhaps reecting the operation of
the foreign tax credit mechanism.
4. Empirical results on taxation and subsidiary location
In this section, we estimate the impact of host and additional parent
country taxation on the choiceof location of a newforeign subsidiary. In
our estimation, we explain subsidiary location choice only in the initial
year of establishment to ensure that observations regarding different
location episodes are independent. We condition on a foreign subsidi-
ary location, and hence examine the impact of international taxation
on the location choice among competing foreign locations.Our sample consists of new foreign subsidiaries that incorporate
in one of the years from 1999 to 2003. The data set encompasses 33
European countries, which implies that for a multinational resident
in a particular home country there are 32 foreign location options. Ac-
cordingly, for each new foreign subsidiary we construct 32 binary
variables that take on a value of one if the actual location is in a cer-
tain foreign country and zero otherwise. Regarding each potential lo-
cation for each new foreign subsidiary, there thus is a binary choice.
Location choice is assumed to be determined by the various countries’
tax rates and a range of other location or country characteristics.9 The
underlying binary choice model is estimated using the conditional
logit approach of McFadden (1974, 1976).
10
In this approach, explana-tory variables vary across alternative outcomes. All regressions include
year xed effects, which are not reported.11 We also control for possible
clustering of the errors at the level of the potentialhostcountry to allow
for the possibility that shocks to the attractiveness of host countries af-
fect more than a single subsidiary location choice.12 Specically, we re-
port heteroscedasticity-consistent standard errors that are adjusted for
clustering at the level of the host country.
In regression 1 of Table 5, we relate foreign subsidiary location only
to the effective tax rate. The number of observations is 26,648 rather
than 29,088 (or 909 new subsidiaries times 32 potential location coun-
tries) due to missing observations of the effective tax variable for some
combinations of countries and years. The effective tax variable enters
with a negative coef cient of −0.87 that is statistically insignicant.
Table 3
Existence of bilateral tax treaties for European country pairs in 2003.
Income From: Austria Belgium Bulgaria Croatia Cyprus Czech
Rep.
Denmark Estonia Finland France Germany Greece Hungary Iceland Ireland Italy
To:
Austria X 1 1 1 1 1 1 1 1 1 1 1 1 0 1 1
Belgium 1 X 1 1 1 1 1 0 1 1 1 1 1 0 1 1
Bulgaria 1 1 X 1 1 1 1 0 1 1 1 1 1 0 1 1
Croatia 1 1 0 X 1 1 1 0 1 1 1 1 1 0 0 1
Cyprus 1 1 1 1 X 1 1 0 0 1 1 1 1 0 1 1
Czech Rep. 1 1 1 1 1 X 1 1 1 1 1 1 1 1 1 1
Denmark 1 1 1 1 1 1 X 1 1 1 1 1 1 1 1 1
Estonia 0 0 0 0 0 1 1 X 1 1 1 0 0 1 1 1
Finland 1 1 1 1 0 1 1 1 X 1 1 1 1 1 1 1
France 1 1 1 1 1 1 1 1 1 X 1 1 1 1 1 1
Germany 1 1 1 1 1 1 1 1 1 1 X 1 1 1 1 1
Greece 1 1 1 1 1 1 1 0 1 1 1 X 1 0 1 1
Hungary 1 1 1 1 1 1 1 0 1 1 1 1 X 0 1 1
Iceland 0 0 0 0 0 1 1 1 1 1 1 0 0 X 0 0
Ireland 1 1 1 0 1 1 1 1 1 1 1 0 1 0 X 1
Italy 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 X
Latvia 0 0 0 1 0 1 1 1 1 1 1 0 0 1 1 0
Lithuania 0 0 0 1 0 1 1 1 1 1 1 0 0 1 1 1
Luxembourg 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 1
Malta 1 1 1 1 1 1 1 1 1 1 1 0 1 0 0 1
Netherlands 1 1 1 1 0 1 1 1 1 1 1 1 1 1 1 1
Norway 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Poland 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1Portugal 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 1
Romania 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1
Russia 1 1 1 1 1 1 1 0 1 1 1 0 1 0 1 1
Slovak Rep. 1 1 1 1 1 1 1 0 1 1 1 1 1 0 1 1
Slovenia 1 1 0 0 1 1 1 0 1 1 1 0 1 0 1 1
Spain 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 1
Sweden 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Switzerland 1 1 1 0 0 1 1 0 1 1 1 1 1 1 1 1
Turkey 1 1 1 1 0 0 1 0 1 1 1 0 1 0 0 1
United Kingdom 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Notes: This table addresses whether a tax treaty was in effect to deal with income received by countries listed in the rows and originating from countries listed in the columns.
Specically, 1 denotes that a bilateral tax treaty was applicable and 0 denotes that a tax treaty was not applicable. The table is not exactly symmetric as the dates of rst application
of a treaty may slightly differ between two treaty partners. Source: International Bureau of Fiscal Documentation and various ministries ’ websites.
9 Note that we explain changes in the rm's structure by the level of international
taxation. Expanding rms may have a need to establish new subsidiaries even if there
are no changes in the level of taxation.
10 The conditional logit model imposes the axiom of independence of irrelevant alter-
natives (IIA), which implies that adding a third option or changing the characteristics
of a third alternative does not affect the relative odds between any two options consid-
ered. As a robustness check, we have estimated a nested logit model that allows us to
relax the IIA assumption yielding results broadly similar to the ones reported in Table 5
(unreported).11 To enable estimation of the year xed effects, we simultaneously consider how a
multinational selects the location of its foreign subsidiaries over the entire period
1999–2003, rather than 1 year at a time.12 The possibility that multinationals' location choices might be correlated both geo-
graphically and over time has been suggested in a number of recent empirical studies.
Crozet et al. (2004), for instance, nd evidence of positive spillovers affecting the loca-
tion choices of French rms investing abroad suggesting that French multinationals
cluster geographically in a gradual process of agglomeration and market penetration.
Head and Mayer (2004) study the determinants of the location choices of Japanese
multinationals in Europe and nd a similar tendency for these multinational rms to
cluster in the same countries.
952 S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
8/13
The corresponding estimate of the marginal effect of the effective tax
rate on the probability of location is−0.22, meaning thata one percent-
age point increase in theeffective taxrate (from a level of zero) reduces
the probability of location by 0.22%.
Next, regression 2 includes a host of non-tax control variables, and
it yields an estimated coef cient for the effective tax rate variable of
−4.29 that is signicant at the 1% level. The corresponding estimate
of the marginal effect of the effective tax rate on the probability of
location is −0.78. Among the controls, a country's relative GDP is es-
timated to increase the probability of subsidiary location with signif-
icance at the 1% level. Contiguous countries are also more likely to
receive foreign subsidiaries with signicance at the 1% level. The dif-
ference in labor costs variable enters with a negative coef cient that
is signicant at 1%, perhaps suggesting that high wages in prospectivehost countries signal sought after labor skills. The economic freedom
and EU membership variables both enter with positive coef cients
that are signicant at 10% and 1%, respectively.13
Regression 3 substitutes the host country corporate tax rate for the
effective tax rate. The estimated parameter on the host country tax var-
iablehas a value of −3.42and it is signicant at the 1%level. Inlinewith
this, a one percentage point increase in the host countrytax rate is esti-
mated to reduce the probability of location by 0.63%. The control vari-
ables enter regression 3 in qualitatively the same way as before.
Regression 4 in turn includes the international tax variable –
reecting both non-resident withholding taxation in the host country
and additional parent country corporate taxation – with an estimated
coef cient of −1.87 that is signicant at 5%. The corresponding mar-
ginal effect of the international tax rate on the probability of location
is estimated to be relatively small in absolute value at 0.19.
Next, regression 5 jointly includes the host country tax rate and
the international tax rate, yielding estimated coef cients of −
4.38and −3.93, that are both signicant at 1%, with corresponding esti-
mated marginal effects of −0.82 and −0.73.
Finally, regression 6 splits up the effective tax rate into its three com-
ponents: the host country corporate tax rate, the non-resident withhold-
ing tax rate, and the additional parent country corporate tax rate.
Parameter estimates for the host country tax rateand parent country cor-
porate tax rate are negative and statistically signicant at 1%, while the
non-resident dividend withholding tax rateobtains a negative coef cient
that is statistically insignicant. A one percentage point increase in the
host country and additional parent country tax rates are estimated to re-
duce theprobabilityof location by 0.90% and1.07% respectively,while the
analogous estimated effect of the non-resident withholding tax is 0.06%.
Our results suggest that host country and additional parent country
taxation both play a signi
cant role in multinationals' location choices.
13 The presence of intangible assets or R&D intensity could similarly affect the sensi-
tivity of international location decisions to taxation. Recent evidence provided by
Dischinger and Riedel (2011) shows that corporate taxation signicantly affects the in-
ternational location of intangible assets given a multinational rm's structure. To test
this, we distinguished between rms that belong to sectors with low and medium or
high R&D intensity using information from the OECD Science, Technology and Industry
Scoreboard (see OECD, 2003). Following the OECD taxonomy, medium or high R&D in-
tensity sectors have R&D expenditure exceeding 2% of total value added. We estimated
separate regressions analogous to regression 2 of Table 5 for the samples of rms be-
longing to the two groups of sectors, yielding very similar parameter estimates for
the effective tax rate variable (unreported).
Latvia Lithuania Luxembourg Malta Netherlands Norway Poland Portugal Romania Russia Slovak
Rep.
Slovenia Spain Sweden Switzerland Turkey United
Kingdom
0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 0 1 1 1 1 1 1 1 0 1 0 1 1 1 0 1
1 1 0 0 1 1 1 0 1 1 1 0 0 1 0 0 1
0 0 0 1 0 1 1 0 1 1 1 1 0 1 0 0 11 1 1 0 1 1 1 1 1 1 1 1 1 1 1 0 1
1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 1 1
1 1 0 0 1 1 1 0 0 0 0 0 0 1 0 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
0 0 1 0 1 1 1 1 1 0 1 0 1 1 1 0 1
0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 0 1 1 1 1 0 0 0 0 1 1 1 0 1
1 1 1 0 1 1 1 1 1 1 1 1 1 1 1 0 1
0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
X 1 0 0 1 1 1 0 1 0 1 1 0 1 1 0 1
1 X 0 0 1 1 1 0 1 0 1 1 0 1 1 0 1
0 0 X 1 1 1 1 1 1 1 1 1 1 1 1 0 1
1 0 1 X 1 1 1 1 1 0 1 0 0 1 1 0 1
1 1 1 1 X 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 X 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 X 1 1 1 1 1 1 1 1 1 10 0 1 1 1 1 1 X 1 1 1 0 1 0 1 0 1
1 1 1 0 1 1 1 1 X 1 1 1 1 1 1 1 1
0 0 1 0 1 1 1 1 1 X 1 0 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 X 1 1 1 1 1 1
1 1 1 0 1 1 1 0 1 1 1 X 1 1 1 0 1
0 0 1 0 1 1 1 1 1 1 1 1 X 1 1 0 1
1 1 1 1 1 1 1 1 1 1 1 1 1 X 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 X 0 1
0 1 0 0 1 1 1 0 1 1 1 0 0 1 0 X 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 X
953S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
http://-/?-http://-/?-http://-/?-http://-/?-
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
9/13
To evaluate the implications of tax rate changes in host and parent coun-
tries for location decisions,one needs to recognize thatsuch changes gen-
erally alter both our host country and additional parent country tax
variables. An increase in the host country corporate tax rate, for instance,
would lead to an increase in the host country tax variable and an
off-setting reduction in the additional parent country tax variable if a
full foreign tax credit is available (this is the case if the host country tax
rate remainsbelow theparent countrytax rate absent non‐resident with-
holding taxes). In this instance, the effective tax rate would remainunchanged, and the implied estimated impact on location would be
small, as the estimated marginal effects of the host and additional parent
country tax variables on location probabilities of −0.90 and −1.07 are
very similar. An increase in the host country corporate tax rate, however,
increases the host country tax variable while leaving the additional par-
ent country tax variable unchanged, if the host country tax rate exceeds
the parent country tax rate. In this instance, a higher host country corpo-
ratetax ratediscourages locationin the host country, while increasing the
combined, effective tax rate. Overall, our estimation is consistent with a
negative relationship between the host country corporate tax rate and
subsidiary location as generally reported in the literature surveyed by
De Mooij and Ederveen (2006).
Our nding of a large role for the additional parent country tax in
foreign subsidiary location is new, and perhaps surprising, as many
countries in the world allow parent country taxes on foreign source
income to be deferred until dividend repatriation, diminishing its po-
tential to affect location decisions. Similarly, some countries allow
so-called worldwide income averaging. This practice allows multina-
tionals resident in, for instance, the U.S. to claim foreign tax credits for
foreign taxes paid in high-tax countries against U.S. taxes due on in-
come from low-tax countries, potentially reducing the burden of par-
ent country corporate income taxation.
The result that foreign subsidiary location is sensitive to additional
parent country taxationreects thatparent country taxationis rather dis-
criminatory, as it only applies to parent rms residing in the pertinent
parent country.Additional parent country taxationthus discouragesmul-
tinational rm ownership of foreign subsidiaries that are subject to addi-
tional parent country taxation. The estimated negative impact of parent
country location on subsidiary location implies that multinationals acton an apparently signicant incentive to bring about an international
ownership pattern of subsidiaries that is internationally tax-ef cient.
Substantial withholding taxes also put particular foreign owners at a
comparative disadvantage at owning local assets vis-à-vis any other for-
eign owners that are lowly taxed and local owners. All the same, wend
that non‐resident dividend withholding taxes are statistically insigni-
cant in determining subsidiary location decisions. To explain this, rst
note that the EU Parent-Subsidiary Directive provides that no withhold-
ing tax shall be levied on dividend payments between related cross-
border companies. This makes the application of a withholding tax rare
in our dataset. Second, companies can potentially avoid withholding
taxes by creating conduit companies or by letting the subsidiary compa-
ny provide the parent company with a loan instead of a dividend.14
Next, Table 6 reports several robustness checks to gain additionalinsight in the impact of the international tax system on foreign
subsidiary location.15 First, we recognize that the calculation of our
effective tax variable is based on countries’ top statutory tax rates,
thus ignoring potential differences in tax base denitions regarding,
for instance, interest deductibility that equally affect the tax burden
in a particular country. As an alternative effective taxmeasure, regres-
sion 1 of Table 6 includes the Effective Average Tax Rate (EATR) for
outbound FDI as computed by ZEW (2008) starting from regression
2 of Table 5. The EATR reects statutory information on tax rates, in-
terest deductibility, as well as allowances for capital depreciation,
and it is constructed as the overall tax liability triggered by a projectdivided by the overall returns generated by the project in the absence
of taxation (see Devereux and Grif th, 1999, 2003). The EATR approx-
imatesan average taxon FDI, unlikeour effective tax rate based on top
statutorytax rates,and thus canbe expected to inuence international
location decisions. In regression 1, the EATR variable obtains a nega-
tive coef cient of −2.51 that is statistically insignicant. The marginal
effect in the EATR regression is −0.60. The ZEW (2008) data do not
provide a split up theEATR variable into host and parentcountry com-
ponents and hence we cannot perform regressions using EATR-like tax
variables analogous to regressions 3–6 of Table 5.
Next, we investigate whether the responsiveness of subsidiary lo-
cation is different for subsidiaries with a low or high return on assets.
Subsidiaries with a low return on assets presumably pay low taxes,
suggesting that there is a reduced role of taxation to affect loca-
tion choices. Alternatively, subsidiary location of foreign subsidiaries
with a high return on assets could be rather insensitive to taxation,
if the high return on assets reects the realization of rents that are lo-
cation and owner specic. To investigate this, columns 2 and 3 of
Table 6 report regressions for the samples of subsidiaries with a rate
of return on assets below and above the sample median. Otherwise,
these two regressions are analogous to regression 6 of Table 5. The
host country tax variable and the additional parent country tax
enter with negative and statistically signicant coef cients in both
regressions 2 and 3. Interestingly, the estimated marginal effects of
the host and additional parent country taxes on the probability of lo-
cation are both smaller in absolute terms for the high-ROA sample,
which suggests that subsidiaries with a high rate of return experience
rents that are location or owner specic.
Next, we consider whether the responsiveness of location is differ-ent for subsidiaries with low and high xed assets. Subsidiaries that
use high xed assets may be dif cult to relocate physically, and
hence one expects their location to be less sensitive to the host coun-
try tax. Regressions 4 and 5 are based on samples of subsidiaries with
ratios of xed assets to total assets below and above the median. In
regressions 4 and 5, the host country tax and the additional parent
country tax both enter with negative and statistically signicant coef-
cients (albeit at different levels of signicance). The implied margin-
al effects of both tax rates on the probability of location are relatively
large in absolute terms for the sample of subsidiaries with low xed
assets. These results suggest that the location of subsidiaries with
low xed assets is relatively more sensitive to host and parent coun-
try taxation.16
As discussed, the option to defer parent country taxes may makeforeign subsidiary location less responsive to the additional parent
country tax. Previously, Huizinga and Voget (2009) have collected in-
formation on whether the deferral option is available for parent rms
located in a particular home country. We can use this information to
check whether the deferral option indeed mutes the responsiveness
of location to additional parent country taxation. To start, regression
6 in Table 6 introduces an interaction of the effective tax and a
14 The conversion of dividend payments into a loan to the parent potentially also
eliminates parent country taxation on repatriated income. Under US CFC rules, howev-
er, such loans could be labelled ‘deemed dividends’ and trigger parent country taxa-
tion. In the EU, transactions of this kind may be subject to transfer pricing rules, but
adjustments of the transfers to qualify as dividends and subsequent adjustments of
parent country taxation seem to be the exception rather than the rule.15 As additional robustness checks, we have performed regressions that inter alia (i)
exclude holding companies, (ii) include intermediate companies that are both parent
and subsidiary, and (iii) include tax variable interactions with a dummy variable sig-
naling a previously established subsidiary in the country. Our main results on the im-
pact of the international taxsystem on foreign subsidiary location are unaltered in each
of these regressions (unreported).
16 We tested for the equality of coef cients on the tax variable across the two regres-
sions that result from dividing the sample into low ROA/high ROA and low xed assets/
high xed assets subsamples as reported in Table 6 using a Wald test. The results indi-
cate that only the withholding tax variable obtains signicantly different coef cients in
each of the two pairs of regressions. In all four regressions, the withholding tax variable
tax variable, however, obtains a coef cient that itself is insignicant.
954 S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
10/13
deferral option dummy in regression 2 of Table 5. The effective tax
variable now obtains a coef cient of −9.17 that is signicant at 1%
and lower than the corresponding coef cient of −
4.29 in regression
2 of Table 5. The interaction of the effective tax with the deferral op-
tion dummy obtains a positive coef cient of 4.85 that is signicant at
1%. The corresponding marginal effects on the probability of location
Table 4
Descriptive statistics for subsidiaries of European multinationals.
Panel A. Number of parent companies and subsidiaries used in basic regression
Country Number of parent companies by home country Number of subsidiaries
By home country By host country
Austria 9 9 6
Belgium 29 36 34
Bulgaria 9 11 2
Croatia 0 0 1Cyprus 0 0 0
Czech Republic 0 0 5
Denmark 26 28 110
Estonia 5 6 1
Finland 19 23 30
France 84 116 36
Germany 56 75 97
Greece 33 48 4
Hungary 0 0 6
Iceland 2 3 4
Ireland 9 11 39
Italy 64 78 38
Latvia 1 2 2
Lithuania 0 0 0
Luxembourg 1 1 2
Netherlands 64 89 51
Norway 33 45 74Poland 5 6 39
Portugal 19 21 11
Romania 0 0 9
Russia 1 1 0
Slovak Republic 1 1 0
Slovenia 6 6 0
Spain 49 59 88
Sweden 69 76 68
Switzerland 13 14 6
Turkey 0 0 0
United Kingdom 115 144 146
Total 722 909 909
Panel B. Summary statistics for variables in basic regression
Variable n Mean Standard dev. Min Max
Subsidiary location 26,648 0.034 0.181 0 1
Effective tax 26,648 0.353 0.073 0.125 0.750International tax 26,648 0.051 0.070 0 0.550
Host country tax 26,648 0.302 0.083 0 0.567
GDP bilateral 26,648 0.033 0.054 0.0003 0.264
Contiguity 26,648 0.166 0.371 0 1
Difference in labor costs 26,648 0.270 0.563 −2.098 2.373
Economic freedom 26,648 6.430 1.021 3.800 8.425
EU membership 26,648 0.464 0.499 0 1
Panel C. Correlation matrix for variables in basic regression
Subsidiary
location
Effective
tax
International
tax
Host
country tax
GDP
bilateral
Contiguity Difference in
labor costs
Economic
freedom
EU
membership
Sub sidiar y loca tion 1 .0 00 0
Effective tax −0.0059 1.0000
International tax −0.0798** 0.3369** 1.0000
Host country tax 0.0628** 0.5976** −0.5536** 1.0000
GDP bilateral 0.1669** 0.2662** −
0.2907** 0.4830** 1.0000Contiguity 0.0733** 0.0794** −0.1098** 0.1638** 0.2324** 1.0000
Difference in labor costs −0.0923* 0.0397* 0.1522* −0.0945* −0.2416 −0.1455 1.0000
Economic f reedom 0 .1 03 0** −0.1442** −0.1667** 0.0144** 0.1943** 0.0794** −0.6670* 1.0000
EU membership 0.1402** 0.0622** −0.4990** 0.4800** 0.4975** 0.2092** −0. 4547** −0.4891* 1.0000
Notes: Subsidiary location is a dummy variable equaling 1 if a subsidiary is located in a country and 0 otherwise. Effective tax is the tax rate on dividend income generated in the potential
subsidiary country resulting from corporate income taxation in host and parent countries and non-resident withholding taxation in the host country. International tax is the difference be-
tweenthe effectivetax andthe host country corporatetax.Host country taxis thecorporate incometax in thesubsidiary country, includinglocaltaxes andpossiblesurcharges.GDP bilateral
is the ratio of the GDP in a potential host country to the sum of GDP of all potential host countries. Contiguity is a binary variable equal to 1 if the parent and potential host countries have a
commonborder. Difference in labor costs is thelog of theratioof labor costs in thehome country andlaborcosts in thepotential host country. Economic freedom is an index scaledbetween
0and10reectingthe followingFraser indicators of thepotential host country: soundnessof legalsystem, absence of trade barriers,and absence of price controls.EU membership is a binary
variable equal to 1 if the potential host country is a member of the European Union. Basic regression refers to regression 2 in Table 5. * denotes signicance at 5%; ** signicance at 1%.
955S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
11/13
are −1.63 and 0.86.These resultssuggestthat the deferral option re-
duces the responsiveness of location to the effective country tax by
about half.
In analogous fashion, regression 7 of Table 6 includes an interac-
tion of the international tax variable with the deferral option
dummy in regression 5 of Table 5. The international tax variable
and its interaction with the deferral option dummy enter with neg-ative and positive coef cients, respectively, that are signicant at
1% and 5% respectively. The implied marginal effects of the two var-
iables on the probability of location are −3.10 and 2.48, which sug-
gests that the deferral option reduces the responsiveness of location
to the international tax variable by 80%. Finally, regression 8 of
Table 6 includes an interaction of the additional parent country
tax with the deferral option dummy in regression 6 of Table 5.
The additional parent country tax obtains a negative coef cient
that is signicant at 5%, while its interaction with the deferral op-
tion dummy obtains a positive coef cient that is statistically insig-
nicant. Estimated marginal effects of the two variables on the
probability of location suggest that the deferral option reduces the
responsiveness of location to the additional parent country tax by
about 72%.
Overall, the robustness checks conrm that international double
taxation, as reected in our international tax and additional parent
country tax variables, is important in explaining location decisions re-
garding new foreign subsidiaries. More specically, our results sug-
gest that foreign subsidiary location is relatively sensitive to
international double taxation for subsidiaries with a low return on as-
sets, low xed assets and in the absence of the deferral option.
5. Conclusions
This paper provides evidence on the implications of international
taxation for the location of the foreign subsidiaries of multinational
rms. Our analysis uses panel data on the structure of multinational
rms in 33 European countries over the period 1999–2003. This rich
data set allows us to estimate the separate effects of host country
corporate income taxation, host country non-resident dividend with-
holding taxation, and additional parent country corporate income
taxation on the location decisions of multinational rms.
Our main result is that the additional parent country corporate
taxation of foreign source income has an independent, strongly nega-
tive effect on the probability of foreign subsidiary location in potential
Table 5
Taxation and foreign subsidiary location. The dependent variable is subsidiary location equaling 1 if a subsidiary is located in a country and 0 otherwise. Effective tax is the tax rate
on dividend income generated in the potential subsidiary country resulting from corporate income taxation in host and parent countries and non-resident withholding taxation in
the host country. Host country corporate tax is the corporate income tax in the subsidiary country, including local taxes and possible surcharges. International tax is the difference
between the effective tax and the host country corporate tax. Withholding tax is the non-resident withholding tax burden imposed by the host country. Additional parent country
corporate tax is the additional corporate tax burden imposed by parent country after double tax relief. GDP bilateral is the ratio of the GDP in a potential host country to the sum of
GDP of all potential host countries. Contiguity is a binary variable equal to 1 if the parent and potential host countries have a common border. Difference in labor costs is the log of
the ratio of labor costs in the home country and labor costs in the potential host country. Economic freedom is an index scaled between 0 and 10 reecting the following Fraser
indicators of the potential host country: soundness of legal system, absence of trade barriers, absence of price controls, and transfers and subsidies as a share of GDP. EU member-
ship is a binary variable equal to 1 if the potential host country is a member of the European Union. All regressions include year xed effects that are not reported. Sample reects
location decisions of new foreign subsidiaries. Estimation is by conditional logit model. Standard errors that are heteroskedasticity-consistent and adjusted for clustering at the levelof the host country are reported between brackets. Marginal effect is the slope of the probability curve with respect to an included tax variable evaluated at zero while other ex-
planatory variables are evaluated at their mean. * denotes signicance at 10%; ** signicance at 5% and *** signicance at 1%.
(1) (2) (3) (4) (5) (6)
Variables (expected sign) Tax
only
Basic
regression
Host country
corporate tax
International
taxation
Host country corporate and
international taxes
Withholding and parent
country corporate tax
Effective tax (−) −0.871 −4.292***
( 0.57 4) ( 0.63 4)
Host country corporate tax (−) −3.424*** −4.376*** −4.806***
(0.500) (0.624) (0.654)
International tax (−) −1.869** −3.927***
(0.938) (1.042)
Withholding tax (−) −0.322
(1.400)
Additional parent country corporate
tax (−)
−5.731***
(1.400)
GDP bilateral (+) 7.420*** 7.196*** 5.404*** 7.470*** 7.620***(0.575) (0.548) (0.475) (0.566) (0.571)
Contiguity (+) 0.396*** 0.372*** 0.372*** 0.395*** 0.402***
(0.084) (0.085) (0.086) (0.084) (0.084)
Differences in labor costs (+/−) −0.692*** −0.768*** −0.573*** −0.705*** −0.737***
(0.179) (0.190) (0.174) (0.186) (0.181)
Economic freedom (+) 0.140* 0.139* 0.291*** 0.134* 0.133*
(0.073) (0.073) (0.067) (0.074) (0.074)
EU membership (+) 0.952*** 1.192*** 0.814*** 0.981*** 1.107***
(0.105) (0.119) (0.117) (0.123) (0.129)
Observations 26,648 26,648 26,648 26,648 26,648 26,648
Pseudo R-squared 0.001 0.135 0.133 0.127 0.136 0.137
Marginal effects of tax variables
Effective tax −0.218 −0.784
Host country corporate tax −0.632 −0.816 −0.895
International tax −0.189 −0.732
Withholding tax −
0.060Additional parent country
corporate tax
−1.070
956 S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
12/13
host countries, despite the fact that parent country taxation can gen-
erally be deferred until income is repatriated. This result may reect
that a parent country's taxation is rather discriminatory as it only ap-
plies to parent rms residing in the parent country. The high sensitiv-
ity of foreign subsidiary location to parent country taxation suggests
that this tax is particularly distortive. Paradoxically, the parent coun-
try tax may be highly distortive on account of the foreign credit
mechanism – aiming to alleviate international double taxation – as
the foreign tax credit mechanism produces a high variability of the
(post credit) parent country tax across foreign location choices. The
sensitivity of foreign subsidiary location to parent country taxation
strengthens the case for abolishing worldwide taxation in favor of
territorial taxation.
Additional parent country taxation may not only distort the for-
eign subsidiary location decision, but also the capital investment de-
cision once location is determined. The impact of parent country
taxation on capital investment abroad would be an interesting area
for future research.
Table 6
Taxation and foreign subsidiary location: robustness tests. The dependent variable is subsidiary location equaling 1 if a subsidiary is located in a country and 0 otherwise. Effective
tax is the tax rate on dividend income generated in the potential subsidiary country resulting from corporate income taxation in host and parent countries and non-resident with-
holding taxation in the host country. Host country corporate tax is the corporate income tax in the subsidiary country, including local taxes and possible surcharges. International
tax is the difference between the effective tax and the host country corporate tax. Withholding tax is the non-resident withholding tax burden imposed by the host country. Ad-
ditional parent country corporate tax is the additional corporate tax burden imposed by parent country after double tax relief. GDP bilateral is the ratio of the GDP in a potential host
country to the sum of GDP of all potential host countries. Contiguity is a binary variable equal to 1 if the parent and potential host countries have a common border. Difference in
labor costs is the log of the ratio of labor costs in the home country and labor costs in the potential host country. Economic freedom is an index scaled between 0 and 10 re ecting
the following Fraser indicators of the potential host country: soundness of legal system, absence of trade barriers, absence of price controls, and transfers and subsidies as a share of
GDP. EU membership is a binary variable equal to 1 if the potential host country is a member of the European Union. All regressions include year xed effects that are not reported.
The EATR is the cross-border Average Effective Corporate Tax Rate taken from ZEW (2008). Deferral is a dummy variable equal to 1 if deferral is possible and zero otherwise. Re-gressions 2 and 3 split the sample into subsidiaries that have respectively a ratio of pre-tax prots to total assets below or equal to and above the median value. Regressions 4 and 5
split the sample into subsidiaries that have respectively a ratio of xed assets to total assets below or equal to and above the median value. Sample re ects location decisions of new
foreign subsidiaries. Standard errors thatare heteroskedasticity-consistent and adjusted for clustering at the level of the host country are reported between brackets. Marginal effect
is the slope of the probability curve with respect to an included tax variable evaluated at zero while other explanatory variables are evaluated at their mean. * denotes signicance at
10%; ** signicance at 5% and *** signicance at 1%.
(1) (2) (3) (4) (5) (6) (7) (8)
Variables (expected sign) EATR Low ROA High ROA Low xed
assets
High xed
assets
Deferral Deferral
two-way
split
Deferral
three-way
split
EATR −2.513
(1.883)
Effective tax (−) −9.166***
(1.798)
Host country corporate
tax (−)
−4.367*** −5.238*** −5.911*** −3.930*** −4.504*** −4.882***
(0.806) (0.912) (0.955) (0.869) (0.622) (0.652)
International tax (−
) −
16.61***(5.620)
Withholding tax (−) 0.383 −1.330 1.036 −1.114 −0.202
(1.725) (2.068) (2.191) (1.627) (1.400)
Additional parent country
corporate tax (−)
−5.951***
(1.802)
−5.594***
(1.982)
−6.803**
(3.239)
−5.277***
(1.475)
−17.850**
(8.167)
Deferral× Effective tax (+) 4.853***
(1.711)
Deferral×International
tax (+)
13.26**
(5.580)
Deferral×Additional parent
country tax (+)
12.86
(8.180)
GDP bilateral (+) 4.765*** 7.066*** 8.094*** 9.367*** 5.980*** 7.365*** 7.478*** 7.616***
(0.810) (0.779) (0.772) (0.811) (0.771) (0.574) (0.561) (0.568)
Contiguity (+) 0.664*** 0.448*** 0.347*** 0.408*** 0.396*** 0.412*** 0.411*** 0.416***
(0.109) (0.114) (0.119) (0.122) (0.113) (0.084) (0.084) (0.084)
Labor cost (+/−) −0.197 −1.027*** −0.353 −1.081*** −0.575*** −0.682*** −0.713*** −0.731***
(0.693) (0.199) (0.300) (0.330) (0.214) (0.178) (0.186) (0.180)Eco nomic free dom (+) 0.038 −0.041 0.351*** 0.048 0.183* 0.149** 0.133* 0.133*
(0.280) (0.080) (0.104) (0.113) (0.094) (0.072) (0.074) (0.073)
EU membership (+) 1.446*** 0.902*** 1.407*** 1.552*** 0.849*** 0.897*** 0.977*** 1.112***
(0.159) (0.168) (0.185) (0.213) (0.158) (0.107) (0.126) (0.131)
Observations 12,175 14,176 12,384 11,658 14,990 26,648 26,648 26,648
Pseudo R-squared 0.108 0.119 0.170 0.192 0.106 0.137 0.137 0.138
Marginal effects of tax variables
Effective tax −0.603 −1.625
Host country corporate tax −1.093 −0.366 −1.342 −0.640 −0.841 −0.910
International tax −3.100
Withholding tax 0.097 −0.094 0.237 −0.174 −0.037
Additional parent country
corporate tax
−1.486 −0.391 −1.545 −0.854 −3.329
Deferral×Included
tax variable
0.860 2.475 2.398
957S. Barrios et al. / Journal of Public Economics 96 (2012) 946 –958
-
8/16/2019 Barrios Et Al. (2012) International Taxation and Multinational Firm Location Decisions
13/13
Appendix A
References
Altshuler, R., Grubert, H., 2001. Where will they go if we go territorial? Dividend ex-emption and the location decisions of U.S. multinational corporations. NationalTax Journal 54, 787–809.
Barrios, S., Görg, H., Strobl, E., 2005. Foreign direct investment, competition and indus-trial development in the host country. European Economic Review 49, 1761 –1784.
Buettner, T., Ruf, M., 2007. Tax incentives and the location of FDI: evidence froma panelof German multinationals. International Tax and Public Finance 14, 151–164.
Crozet, M., Mayer, T., Mucchielli, J.-L., 2004. How do rms agglomerate? A study of FDIin France. Regional Science and Urban Economics 34, 27–54.
De Mooij, R., Ederveen, S., 2006. What a difference does it make? Understanding theempirical literature on taxation and international capital ows. Economic PapersNo. 261, European Commission. D-G for Economic and Financial Affairs, Brussels.
Desai, M., Foley, C., Hines, J., 2001. Repatriation taxes and dividend distortions. NationalTax Journal 54, 829–851.
Desai, M.A., Hines, J.R., 2002. Expectations and expatriations: tracing the causes andconsequences of corporate inversions. National Tax Journal 55, 409–440.
Devereux, M.P., Grif th, R., 1998. Taxes and the location of production: evidence from apanel of US multinationals. Journal of Public Economics 68, 335–367.
Devereux, M.P., Grif th, R., 1999. The taxation of discrete investment choices. WorkingPaper No. W98/16. Institute for Fiscal Studies, London.
Devereux, M.P., Grif th, R., 2003. Evaluating tax policy for location decisions. Interna-tional Tax and Public Finance 10, 107–126.
Devereux, M.P., Maf ni, G.,2007. The impact of taxation on the location of capital,rmsand prot: a survey of empirical evidence. Oxford University Centre for Business
Taxation Working Paper 07/02.Dischinger, M., Riedel, N., 2011. Corporate taxes and the location of intangible assets
within multinational rms. Journal of Public Economics 95, 691–707.Egger, P., Loretz, S., Pfaffermayr, M., Winner, H., 2009. Bilateral effective tax rates and
foreign direct investment. International Tax and Public Finance 16, 822 –849.Görg, H., Greenaway, D., 2004. Much ado about nothing? Do domestic rms really ben-
et from foreign direct investment? World Bank Research Observer 19, 171 –197.
Hartman, D.G., 1985. Tax policy and foreign direct investment. Journal of Public Eco-nomics 26, 107–121.
Head, K., Mayer, T., 2004. Market potential and the location of Japanese rms in theEuropean Union. Review of Economics and Statistics 86, 959–972.
Hines, J.R., 1996. Altered states: taxes and the location of foreign direct investment inAmerica. American Economic Review 86, 1076–1094.
Huizinga, H., Voget, J., 2009. International taxation and the direction and volume of cross-border M&As. Journal of Finance 64, 1217–1249.
Huizinga, H., Laeven, L., Nicodème, G., 2008. Capital structure and international debtshifting. Journal of Financial Economi