Transcript www.ifs.org.uk

The Impact and Design of
Business Taxation in a Globalised
World
London April 27, 2009
Worldwide Corporate Income Taxes
Kevin S. Markle and Douglas A. Shackelford
University of North Carolina
April 27, 2009
U.S. Presidential debate, September 26, 2008
McCAIN:
“Right now, American business pays the second-highest
business taxes in the world, 35 percent. Ireland pays
11 percent.
“Now, if you're a business person, and you can locate any place
in the world, then, obviously, if you go to the country where
it's 11 percent tax versus 35 percent, you're going to be able
to create jobs, increase your business, make more investment,
et cetera.
“I want to cut that business tax. I want to cut it so that
businesses will remain in the United States of America and
create jobs.”
3
U.S. Presidential debate, September 26, 2008
OBAMA:
“Now, John mentioned the fact that business taxes
on paper are high in this country, and he's
absolutely right. Here's the problem: There are so
many loopholes that have been written into the
tax code, oftentimes with support of Senator
McCain, that we actually see our businesses
pay effectively one of the lowest tax rates
in the world.”
4
What Do We Do?

Estimate effective tax rates (ETRs) using financial
statement information

Compare ETRs for domestics and multinationals

Compare ETRs across countries

Compare ETRs across years

Measure the impact of foreign subsidiaries on ETRs
5
What Do We Find?

Multinationals and domestic firms face similar ETRs.

Average ETR decline from 1988-2007 was 7 percentage points
(20%).

Country ETR order remains constant over time.

Japan has the highest ETRs

U.S. and European countries have above-average ETRs.

Middle East, Tax Havens and Asian (ignoring Japan) countries
have below-average ETRs.
6
Regression Equations
Three specifications:
7
Variables

Coefficients of Interest

β0 = domestic ETR

(β0 + β1) = multinational ETR

ETR = book ETR (from the financial statements)

Numerator is total tax expense (≥0)


Same conclusions using current income tax expense
Denominator is NIBT (>0), robust to other income measures
Controls

Industry (two-digit NAICS)

Year

Size – percentile rank of sales, assets, equity

8
Countries
Sample:
parents in 85 countries
subs in 195 countries
BUT only know sub locations in 2008
Countries
•
•
•
•
•
•
•
•
•
9
Australia**
Canada
China
France
Germany
India
Japan
UK
US
Groups
•
•
•
•
•
•
•
Asian Tigers
Tax Havens
Africa
Asia
Europe
Latin America
Middle East
UK Tax Rates
30
28
25
26
27 27
26 26
24
22
20
25
23
23
24
15
10
5
2003-2007
ETR from f/s
ESTIMATE
ETR =
Cur tax/NIBT
UK Domestic
Manufacturing
Low R&D
UK Multinational
1988 - 2007
ESTIMATE
U.S. Tax Rates
30
30
30
28
25
28
26 27
26
24 24
24
23
20
21
15
10
5
2003-2007
ETR from f/s
ESTIMATE
ETR =
Manufacturing
Cur tax/NIBT
US Domestic
Low R&D
US Multinational
1988 - 2007
ESTIMATE
-
Domestic
Multinational
JAPAN
GERMANY
EUROPE
FRANCE
UNITED STATES
UNITED KINGDOM
AFRICA
LATIN AMERICA
AUSTRALIA
INDIA
CANADA
ASIA
CHINA
ASIAN TIGERS
TAX HAVENS
MIDDLE EAST
2003-2007 ETR by Parent Country
40
35
30
25
20
15
10
5
JAPAN
GERMANY
EUROPE
FRANCE
UNITED STATES
UNITED KINGDOM
AFRICA
LATIN AMERICA
AUSTRALIA
INDIA
CANADA
ASIA
CHINA
ASIAN TIGERS
TAX HAVENS
MIDDLE EAST
ETR conditional on location of subs
40
35
30
25
20
15
10
5
Multinational Tax rates over Time
60
50
40
30
20
10
JAPAN
UNITED STATES
EUROPE
GERMANY
UNITED KINGDOM
TAX HAVENS
AUSTRALIA
FRANCE
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
-
Multinational Gap over Time
60
50
JAPAN dom
JAPAN mn
40
UK dom
30
UK mn
US dom
20
US mn
10
2006
2004
2002
2000
1998
1996
1994
1992
1990
1988
-
Impact of Sub Location by Parent Country
1.5
1.0
0.5
(0.5)
(1.0)
(1.5)
(2.0)
Parents
UK
US
EUROPE
UK
EUROPE
GERMANY
CHINA
JAPAN
TIGERS
ASIA
TAX
HAVENS
Subs
US
(3.0)
LAT
AMERICA
(2.5)
Future Work--Clusters

Companies appear to cluster among countries


e.g., If anywhere in Europe, then in Ireland, the
Netherlands, and Switzerland?
Future work:
How does this clustering affect our
understanding of the taxes on
multinationals?
17
What Do We Find?

Multinationals and domestic firms face similar ETRs.

Average ETR decline from 1988-2007 was 7 percentage points
(20%).

Country ETR order remains constant over time.

Japan has the highest ETRs

U.S. and European countries have above-average ETRs.

Middle East, Tax Havens and Asian (ignoring Japan) countries
have below-average ETRs.
18
The Impact and Design of
Business Taxation in a Globalised
World
London April 27, 2009
K. Markle and D. Shackelford
Comments by
Julian Alworth
20
•
•
•
•
•
•
Descriptive study: Provide a worldwide comparison of
Effective Tax Rates measured from company financial
statements
Despite very significant changes in company tax rates over
three decades country rankings have not changed
significantly
Differences in ETRs between purely domestic companies
(operating in only one jurisdiction) and multinationals are
generally much smaller than expected
Other controlling factors (industry, size, assets etc.) do not
affect the value of ETRs. Interpretation: cross-sectional
(country) effects are the only factors of significance
Companies with subsidiaries in tax haven countries have
lower ETRs but the coefficient is modest
Authors are cautious in drawing conclusions
21

Sometimes difficulty following the language

Multinational data: longer time series/ domestic data poor
especially for European countries
◦ Presumably always speaking of overall consolidated tax expense
of companies but sometimes authors refer to parent taxation
◦ Domestic companies: small samples relative to MNS in some
instances very few observations (France no data –Germany two
years)
– Weakness of data should be stressed more in paper (appendix?)


What are the authors expecting? Are the results
“reasonable” or do they comply with their expectations?
What is the benchmark?
Useful to have a comparison with statutory rates and other
measures of effective rates (Devereux et al. EMTR)
22

Ranking of statutory / ETR similar
◦ Is this expected?

ETR significantly lower than statutory rates
◦ Is this expected?
23
Statutory Rate
Effective rate
Japan
40
1
42
1
United States
39
2
28
2
Germany
39
3
26
3
Canada
36
4
22
8
India
35
5
24
6
France
34
6
25
5
China
33
7
21
9
Australia
30
8
24
6
United Kingdom
30
8
26
3
24
40%
35%
30%
25%
20%
15%
10%
5%
0%
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Statutory
Devereux-EATR
K-S ETR
25
70%
60%
50%
40%
30%
20%
10%
0%
1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
Statutory
Devereux-EATR
K-S ETR
26
MEASURED ETR vs STATUTORY TAX RATE
45
JAPAN
40
35
y = 0.8831x - 4.5859
ACTUAL ETR
30
R² = 0.2776
US
UK
25
GERMANY
20
15
10
5
0
0
5
10
15
20
25
30
35
40
45
STATUTORY RATE
27


There are divergences between accounting
and taxable income.
What are the source of these divergences?
◦
◦
◦
◦

Measurement problem
Temporary vs. permanent differences
Shelters?
Transfer pricing: surprising that data do not show
significantly lower ETRS for MNCs
Testing equation for further work: dependent
variable ETR- Statutory Rate
28

Germany: Multinational companies appear to
have higher ETR than purely domestic
companies
◦ This contrasts with other studies which have
examined financial statement data
◦ Weichenrieder on Germany (1996); Weichenireder
and Mintz (2009) show income shifting out of
exemption country
29
The Impact and Design of
Business Taxation in a Globalised
World
London April 27, 2009