1 Introduction
Globalization is proceeding at a rapid pace. It is therefore not surprising that several so-
cial sciences continue to explore its various implications. Public economists are particularly
interested in how global economic integration aects public policies; consequently, a large
literature on this question has emerged over the years within the eld of Public Economics.
Several authors argue that deepening economic integration may lead to the dissolution of
the welfare state and cause social and political instability (Rodrik, 1997; Rudra, 2002). On
the one hand, economic integration might increase international tax competition and thus
cause lower tax rates on mobile rms (Lee and McKenzie, 1989; Devereux et al., 2008). As a
consequence, the tax burden could shift from mobile (capital) to immobile (labor) production
factors, and fewer resources might be available for redistributive programs. On the other hand,
globalization is often associated with an increase in the cross-border mobility of individuals,
which is expected to additionally strain the welfare state. First, because the wealthy and
highly qualied can more easily defy high personal tax rates through emigration; second,
because generous redistributive transfers might attract the poor and less qualied (Sinn,
2003).
The empirical implication of these arguments is that tax revenues and the amount of
redistributive transfers will decline with deepening globalization. But even though these
theoretical arguments are persuasive, the available empirical evidence is ambiguous. Shelton
(2007) nds that trade openness has had no eect on total government expenditures for a
panel of 100 countries over the 1970-2000 period.1 Slemrod (2004) shows that there is a
negative correlation between corporate tax rates and openness, even though total tax receipts
seem to have been unrelated to openness. Rodrik (1998) and Ram (2009) nd that more open
countries have higher government expenditures.2
This empirical ambiguity is to some extent expected given that globalization may have
conictive eects on scal policy. Therefore, the individual eects of globalization could
cancel each other out at the level of total government expenditures. In response to this
problem, several authors have used less aggregated data, and explored whether globalization
has had an eect on the composition of the public budget. Dreher et al. (2008) study in a
notable contribution this question with a panel of 108 countries over the 1970-2001 period.
However, they nd that the composition of public expenditures has not been aected by
globalization.
1 Note that the terms globalization and trade openness are used interchangeably in this literature. Glob-
alization is, however, a broader concept than trade openness, and additionally implies, inter alia, a global
convergence of cultures and political systems. For a more detailed discussion of the proper meaning of global-
ization see Bhagwati (2004) and Dreher et al. (2008).
2Alesina and Wacziarg (1998), however, present evidence that the positive correlation between openness
and government size is driven by the fact that smaller countries have a larger public sector and are also more
open than larger countries. Benarroch and Pandey (2008) oppose the causal conclusions in Rodrik (1998) by
arguing that his models suer from a reversed causality problem.