The Impact of Taxation on the Location of Capital, Firms and Profit: a survey of empirical evidence Michael P. Devereux University of Warwick
Jan 08, 2016
The Impact of Taxation on the Location of Capital, Firms and Profit:
a survey of empirical evidence
Michael P. Devereux
University of Warwick
Plan of talk
• Introduction to ETPF Research Programme
• Setting the scene for Phase 1 research:– Previous economic research on the impact of
taxes on flows of capital, firms and profit
ETPF Aims and Objectives
1. Undertake high quality economics research on taxation and international business, with European researchers
2. Publicise research results
3. Stimulate research activity – especially empirical - in this area
ETPF: Initial 3 phases of research
1. Positive Economics How do real and financial international flows
respond to taxes?
2. Welfare Economics How does inward investment affect welfare of
domestic residents ? Do governments compete with each other?
3. Tax Design What would be the effects of alternative forms
of taxation?
Phase 1: Capital flows and investment
1. Taxes and the size of the foreign-owned capital stock: which tax rates matter?
– Michael Devereux, Ben Lockwood
2. The impact of non-profit taxes on foreign direct investment: evidence from German multinationals
– Thiess Buettner, Georg Wamser
Phase 1: Financial policy and the location of profits
3. Capital structure and international debt shifting in Europe
– Harry Huizinga, Luc Laeven, Gaetan Nicodeme
4. Profit shifting the EU: evidence from Germany
– Alfons Weichenrieder
Phase 1: Other Projects
5. A survey of existing evidence of the impact of taxes on flows of capital, firms and profits
– Michael Devereux, Giorgia Maffini
6. Corporate income taxes in the EU: an economic assessment of the role of the ECJ
– Stephen Bond, Malcolm Gammie, Socrates Mokkas
Phase 2: Corporation tax and labour
1. Do higher corporate taxes affect the demand for labour, and hence unemployment?
– Ruud De Mooij, University of Rotterdam
2. How far are corporation taxes passed on in lower wages?
– Michael Devereux and Giorgia Maffini, University of Warwick
Phase 2: Corporation tax revenue and tax competition
3. Is there evidence of competition over non-profit taxes paid by business?
– Michael Pfaffermayr, University of Innsbruck
4. Why have corporation tax revenues remained so high, when tax rates have fallen so far?
– Clemens Fuest, University of Köln
Tax Rates v Tax Revenues
20.0%
25.0%
30.0%
35.0%
40.0%
45.0%
50.0%
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
aver
age
stat
utor
y ra
te
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
aver
age
corp
orat
ion
tax
as %
of G
DP
average statutory rate
average revenue as % of GDP
Phase 2: CCCTB Proposal
5. Can a common consolidated corporate tax base reduce variation in effective tax rates across EU?
– Michael Devereux, University of Warwick
Survey of previous Phase 1-type empirical work
• Very little research before 1990
• Most research uses US data– and sponsored by NBER or ITPF
Big issues for empirical research on effects of taxes on behaviour
1. What should be examined?– what is important for policy?
2. What level of data is available?– aggregate data; consolidated data on companies; data
on affiliates
3. What are the appropriate measures of taxation?– marginal or average; forward-looking or backward-
looking
What should be examined (1)?
1. Total Capital StockWhat capital flows are important?
• Foreign direct investment (FDI)?– Financial flows only; and include acquisitions
• Foreign portfolio investment (FPI)?
What should be examined (2)?
2. Multinationals
Do multinationals generate economic benefits?– higher productivity, higher wages, positive
effects on domestic firms?
If so, then their activities should be studied:– Where do they locate?– How much do they invest?– How much tax do they pay?
Data on capital flows,profit and finance
• Aggregate FDI• Bureau of Economic Analysis / US Treasury
– Data on foreign affiliates of US multinationals by country / industry
– tax return data (US Treasury);
• Consolidated financial accounts– Geographic segment of Compustat / Datastream
Microeconomic data
• Unconsolidated data on affiliates, with ownership – Individual-level BEA data (US)– Bundesbank (Germany)– Amadeus (Europe)
• Permits more detailed analysis of individual decisions
Which effective tax rate ?
1 LOCATION CHOICES – Firms choose option with highest post-tax profit
– tax measured by an effective AVERAGE tax rate
2 LEVEL OF INVESTMENT – Firms invest up to marginal project which breaks even
– tax measured by an effective MARGINAL tax rate
3. LOCATION OF PROFIT– Firms can reduce tax liabilities by locating profit in low
tax jurisdictions
– tax measured by STATUTORY tax rate (and possibly withholding taxes)
Measuring effective tax rates
Forward-looking• Based on hypothetical investment project• Can measure AVERAGE and MARGINAL• Only broad tax parameters typically used
Backward-looking• Based on data on tax payments and profit• Using firm level or aggregate data• Depend on past and current decisions
Non-profit taxes ?
Largely ignored in previous literature
• For good reasons? – if sales taxes passed on to consumers– if income taxes borne by labour– then they should not affect location and
investment decisions
• Measurement
EMTR EATR Average Tax Rate
Other
Time series and panel FDI
Slemrod (1990) Hartman (1984) Boskin, Gale (1987) Newlon (1987) Young (1888) Murthy (1989)
Jun (1994)Billington
(1999)
Bilateral FDI Devereux, Freeman (95)
Buettner (2002)Gorter, Parikh (2003)Bénassy Quéré et al
(2005)
Bénassy Quéré et al (2005)
Bellak, Leibrecht (2005)
Buettner (2002)Gorter, Parikh (2003)Bénassy Quéré et al
(2005)
Wei (2000)
Aggregate assets of US multinationals
Grubert, Mutti (1991)Hines, Rice (1994)Swenson (1994)Mutti, Grubert (2004)
Wheeler, Mody (1992)
Hines (1996)
Individual data on multinationals
Devereux, Griffith (98)
Kemsley (1998)Stöwhase (2002)
Individual data on affiliates
Cummins, Hubbard (95)
Buettner, Ruf (2006)
Altshuler et al (2001)Grubert, Mutti (2000)Altshuler, Hubbard
(2003)
Desai et al (2004b)
Broad results on location of capital
• Taxes do have some effect– But estimates vary considerably; cannot
summarise all results into a single elasticity
• For future research:– Need to distinguish more carefully impact of
different elements of tax on different types of decision
– Need to compare across countries
Today’s papers
Examine:
1. Impact of corporation tax on aggregate investment in host countries by US multinationals
• Differentiate impact of effective average tax rate (for location choice) and effective marginal tax rate (for investment)
2. Role of non-profit taxes on location and investment decisions of German multinationals
Data: Aggregate value-added
Aggregate data on affiliates
Firm-level Affiliate level
Profitability Bartelsman, Beetsma (2003)
Jenkins, Wright (1975)
Grubert, Mutti (1991)Hines, Rice (1994)
Klassen et al (1993)Harris (1993)Harris et al (1993)Collins et al (1998)Demirgüc-Kunt,
Huizinga (2001)
Grubert et al (1993)Huizinga, Laeven (2005)
Repatriation policy
Kopits (1972) Kopits (1976) Mutti (1981)Hines, Hubbard (1990)Altshuler, Newlon (1993)Altshuler et al (1995)Altshuler, Mintz (1995)Grubert (1998)Desai et al (2001)Altshuler, Grubert (2004)Desai et al (2006b)
Debt Collins, Shackelford (1992)
Froot, Hines (1995)Altshuler, Mintz
(1995)
Desai et al (2004a)
Other Desai, Hines (1999) Stöwhase (2002) Grubert, Slemrod (1998)Desai et al (2006a)Clausing (2001)
Observed prices
Bernard and Weiner (1990)Swenson (2001b)Clausing (2003)
Broad results on financial policy and location of profit
• Again, taxes have some effect– On rate of profit and financial behaviour– But, again, estimates vary considerably
• For future research– Broadly, research has been more careful in
developing specific hypotheses and tests– But limited by available data: opportunity has
arisen to examine impact of tax in different settings, across countries
Today’s papers
Examine
1. whether rates of profit in German parents and subsidiaries are correlated with tax rates in other countries
2. the extent to which host country tax rates affect the use of debt in European affiliates of multinationals