TAX COMPETITION: THEORY AND
EMPIRICAL EVIDENCE
Michael P. Devereux
Centre for Business Taxation
University of Oxford
copyright rests with the author
Plan
• A brief introduction to the theory
– should we expect competition?
– should we expect a “race to the bottom”?
– can we distinguish “beneficial” and “harmful”
tax competition
• Evidence of tax competition
– Trends in tax rates and revenues
– Econometric evidence
• Conclusions
Basic tax competition theory
•
•
•
Capital mobile across countries, but
labour immobile
Governments provide a public good paid
for by a source-based tax on the return to
capital
They choose a tax rate to reflect
(a) benefits of higher public good provision
(b) loss of capital abroad
•
Tax rate lower than in a closed economy
Questionable assumptions
• Labour not mobile?
• No other taxes available to governments?
• Labour income tax
• VAT
• Residence-based capital income tax
• No imperfect competition, economic rent,
discrete choices
• No publicly-provided goods for production
Incidence & Some Implications
• Taxes on capital (in small open economy)
cannot reduce the post-tax rate of return to
owners
• So are effectively borne by domestic
residents
• Better to tax them directly and avoid
distortion to location of capital
– ie. better NOT to tax capital income
Extensions to model (1)
Competition over discrete location choices,
where firms earn economic profit
1. If firms want to locate near market, large
countries attract investment, though they
may have to pay a subsidy
2. If firms want to locate away from their
competitors, governments can raise
(some) tax without distorting investment
Extensions to model (2)
• 3 levels of decision:
– Where to locate a new facility – depends on average
tax rate
– How much to invest – depend son marginal tax rate
– How much profit to shift to lower-taxed countries –
depends on statutory tax rate
• Governments could compete over any of these 3
tax rates
– should depend on mobility of firms v capital v profit
Is competition harmful or
beneficial? (1)
Compared to what ?
– Closed economy ?
– Partially co-ordinated group of countries ?
• eg. in capital taxes, but not labour taxes
– Fully co-ordinated group of countries ?
– Countries globally co-ordinated ?
Is competition harmful or
beneficial? (2)
• Competition may be generally good, but
– taxes not like ordinary markets; and governments
provide goods that the private sector cannot
– Is competition over environmental pollution
beneficial?
– Competition over only some taxes distort choice of
instruments
• Is a distinction based on competition for firms &
capital as opposed to competition for profit ?
Is competition harmful or
beneficial? (3)
• All competition acts as a constraint in national
policy setting
• If governments act in national interest then real
harm is where other countries hurt, eg.
– global pollution
– preventing other countries raising taxes on capital ?
• But should we tax source-based profit anyway?
– arguably not on efficiency grounds; and can use other
instruments for equity ?
Empirical Evidence
Do governments compete? Some possible
sources of evidence:
1. Trends in tax rates, and tax reforms
2. Evidence of impact of taxes on business
3. Evidence of impact of foreign taxes on
domestic taxes
2004
2003
2002
2001
2000
1999
1998
1997
median
1996
1995
1994
1993
1992
1991
1990
1989
1988
25%
1987
1986
1985
1984
1983
1982
Average OECD Statutory Corporation Tax Rates
55%
50%
45%
40%
35%
30%
unweighted mean
20%
Statutory corporation tax rates:
Old and new member states, 1995-2005
40.0
38.0
36.0
34.0
32.0
30.0
28.0
26.0
24.0
22.0
20.0
1995
1997
1999
15 old member states
2001
2003
10 new member states
2005
Corporation Tax Rate Reductions in EU, 2003-5
Reduction (%)
Year of reform
Austria
34 to 25
2005
Belgium
39 to 33
2003
Cyprus
25 to 15
2003
31 to 28 to 26
2004, 05
Estonia
26 to 24
2005
France
35.4 to 33.8
2005
Greece
35 to 32
2005
Hungary
18 to 16
2004
Italy
36 to 34 to 33
2003, 04
Latvia
22 to 19 to 15
2003, 04
34.5 to 31.5
2005
28 to 27 to 19
2003, 04
Portugal
30 to 25
2004
Slovakia
25 to 19
2004
Czech Republic
Netherlands
Poland
d
ark
Denm
Finla
n
en
gal
Swed
Portu
e
Germ
a
ny
It aly
Malt a
n
ium
Spai
Belg
Fran
c
d Kin
gdom
Luxe
mbou
rg
Net h
erlan
ds
Gree
ce
Unite
nia
publi
c
Slove
h Re
Czec
Aust
ria
Lat vi
a
Lit hu
ania
Hung
ary
Pola
nd
Slova
kia
Esto
nia
Irelan
d
us
Cypr
EU Statutory corporation tax rates, 2005
45
40
35
30
25
20
15
10
5
0
But do trends tell us anything ?
• An implicit hypothesis that (a) globalisation
and hence (b) competition have been
increasing, but
– also requires evidence of pattern of increased
mobility
– theory says little about competition with
imperfect mobility
– tax rates may have moved for other reasons
– tax revenues tell a different story
Outward FDI: US and UK
250
2000 $bn
200
150
100
50
0
1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003
Year
UK
US
But do trends tell us anything ?
• An implicit hypothesis that (a) globalisation
and hence (b) competition have been
increasing, but
– also requires evidence of pattern of increased
mobility
– theory says little about competition with
imperfect mobility
– tax rates may have moved for other reasons
– tax revenues tell a different story
1.0%
0.0%
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
OECD Corporation Tax Revenues as % of GDP
1965-2004
4.5%
4.0%
3.5%
3.0%
2.5%
2.0%
1.5%
GDP weighted average
unweighted average
0.5%
Implicit Tax Rates in Capital: Old and New Member
States 1995 - 2003
35
%
30
25
20
15
10
1995
1996
1997
1998
1999
15 old member states
2000
2001
2002
10 new member states
2003
2. Evidence of impact of taxes on
business behaviour (1)
• Plenty of empirical evidence of effect of
taxes, affecting
– location of firms
– direct flows of capital
• Studies use a variety of measures of both
capital and tax rates
– and hence estimates of elasticities vary widely
2. Evidence of impact of taxes on
business behaviour (2)
• Also evidence of effects of tax on the
location of profit, eg:
– Repatriation of dividends to parent companies
– Use of debt in high-tax subsidiaries
– Transfer prices
– Comparison of profit across countries
Corporate Taxable Income as % GDP,
relative to Statutory Tax Rates, 2004
Corporate Taxable Income as
% of GDP
40.0
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0
10
15
20
25
30
Statutory Tax Rate
35
40
But evidence of tax competition ?
One more stage required before
governments should respond to
concerns of effects of tax:
• What are the welfare consequences of
the induced behaviour of firms? eg:
– Is the aggregate capital stock lower?
– Are productivity and wages lower?
•
Relatively little research on these issues
3. Evidence of impact of foreign
taxes on domestic taxes
Direct examination of relationship between
tax rates
• very little research
• difficulty in identifying appropriate tax rates
– eg. implicit rates (or revenue/GDP) may show
common movements due to correlation in
economic cycle across countries
One study
Devereux, Lockwood & Redoano (2005)
• examine effective marginal tax rate and statutory
tax rate in OECD countries
• find a significant effect on the statutory rate of
statutory rate in other countries
– Consistent with competition for firms (via the effective
average tax rate)
– Consistent with competition for profit
• Overall, results suggest significant competition,
which more than explains reforms up to 2000
Conclusions (1)
Governments do compete in statutory
corporation tax rates
• EU statutory rates of corporation tax have
fallen steadily, and go on falling
– new member states increasing competition
• But revenues have remained buoyant; can
this continue?
Conclusions (2)
Does it matter?
• Depends on what is the “optimal” rate of
source-based corporation tax
– arguably it is zero anyway
– harmonisation of CT may lead to competition in
public services used in production
– if harmonisation were possible, why not
consider residence-based tax within the EU?
Or are political considerations paramount?
Conclusions (3)
• Longer term issues arise with increasing
labour mobility
– less easy to rely on taxes on residents if they
are more mobile
– may eventually imply need to restrict labour
mobility, or agreement on income tax rates