Transcript Tax havens and tax competition Bocconi University, June 18

Tax havens and tax competition
Bocconi University, June 18-19 2007
Setting the research agenda:
What empirical tests would distinguish among
theoretical models?
Federico Revelli
University of Torino
From theory to empirics:
2 main issues
1) Model specification issue:
What are the distinguishing features of alternative tax competition
theoretical models?
In particular, what is the role of the tax system?
2) Econometric issue:
Can common “environmental” shocks be separately identified
from substantive patterns of fiscal competition?
1) Model specification issue
Most existing empirical tax competition literature:
a)
Tax base determination equation: elasticity of mobile tax
bases to (corporate) tax rates: bit = b(it,…,fi, ht)
b)
Tax setting equation: determinants of (corporate) tax rates,
with two variants:

implicit test of tax competition: “internal” determinants only
it = (xit,…,fi, ht)

explicit test of tax competition: tax reaction function, including
the tax rates of “competing” countries on the rhs
it = (nt,…,fi, ht)
n  i
Origins of that empirical specifications:
Pigou tradition
benevolent governments
& refinements
“constrained” governments
Public Choice tradition
Leviathan governments
Role of the tax system: revenue-raising
Unifying approach putting the 2 views in one framework:
“policy-makers are neither wholly benevolent nor wholly self-serving”
Edwards and Keen (1996)
Still, the role of the tax system: revenue-raising
Tax competition in an incomplete information setting
Model with mobile resources and incomplete information
(policy-makers’ quality, competence, honesty unknown)
 political process: selection and discipline
(Besley and Smart, 2007)
Role of the tax system:
1) revenue-raising  tax competition (location)
2) signaling  yardstick competition (reputation)

key issue in empirical tax competition research:
modeling and identifying the two roles of the tax system
Empirical implications:
• “world of multiple taxes” (Desai, Foley and Hines, 2004) and
multiple signals
(e.g., direct & indirect taxes in tax base equation)
• “optimal tax mix” (on mobile and immobile factors)
(Haufler, Klemm and Schjelderup, 2006)
(e.g., testing the backstop hypothesis in tax setting equation)
• “political environment”
(e.g., institutions, constraints, incentives: Janeba and Schjelderup,
2006; Bordignon, Cerniglia and Revelli, 2003)
2) Econometric issue
it = (nt,…,fi, ht)
n  i
a)
“international reflection” problem: reaction to a common
changing environment versus interaction due to competition
(Devereux, Lockwood and Redoano, 2004);
b)
some assumptions need to be imposed either on the pattern of
interaction or on the pattern of environmental shocks;
c)
scepticism about the “reduced-form” tax reaction function
approach;
d)
need an empirical representation “in structural form”
(Ghinamo, Panteghini and Revelli, 2007), modelling tax setting
along with the behavioural responses to taxation.
Ghinamo, Panteghini and Revelli (2007)
• Related literature: empirical tests of tax competition at
international level (Devereux, Lockwood, Redoano, 2002; Slemrod,
2004; Winner, 2005; …)
• Aim is to add to that literature in 2 ways:
1) impact of “volatility” on corporate tax setting policy;
2) move towards a more “structural” approach
• Theoretical framework (Panteghini and Schjelderup, ITPF, 2006):
option pricing model where increased volatility lowers
equilibrium tax rates
• Volatility variables: real interest rate, exchange rate, GDP, plus
“political uncertainty”
• Sample of 51 countries for about 20 years (1983-2003)
Reduced form model:
(1)
τit = ρτit-1 + xit′β + vit′φ + fi + ht + εit
where:
τit = top statutory corporate income tax rate
xit = determinants of corporate tax setting policy (including a
measure of “openness”)
vit′ = various measures of economic and political volatility
Structural form model:
(1)
τit = ρτit-1 + αbit + xit′β + fi + ht + εit
(2)
bit = vit′φ + zit′ + κτit-1 + gi + mt + ηit
where:
τit = top statutory corporate income tax rate
bit = flow of FDI to country i in tear t
vit′ = various measures of economic and political volatility
Main findings
1.
economic volatility (interest rate and exchange rate, but not
GDP) discourages FDI inflow and exerts a negative effect on
corporate tax rates;
2.
political volatility has little effect;
3.
“openness” has a positive effect on the inflow of FDI and a
negative effect on tax rates: taken together (in a reduced form)
these effects tend to cancel each other out.