and bilateral double tax treaties for 26 OECD countries over the period 1991 to
2002. Accordingly, the time-variant tax components vary at the parent-to-host
country-pair rather than the host country (unilateral) level. Overall, we obtain
about 8000 data points for each bilateral tax component. To the best of our
knowledge, this represents the largest existing panel data set of tax compo-
nents. We assess the impact of all tax parameters separately on real bilateral
outbound FDI stocks. The majority of the estimated parameter signs of the
taxation variables is in accordance with the theoretical hypotheses. An omission
of withholding tax rates and depreciation allowances in empirical specifications
seems to ignore important dimensions of taxation and likely leads to biased
estimates of the pure corporate tax effect.
The remainder of the paper is organized as follows. Section 2 provides
a brief review of the existing literature. Section 3 presents a Markusen-type
knowledge-capital model of trade and multinational firms, which accounts for
the mentioned parameters of taxation and the methods of double taxation relief.
Section 4 discusses the major testable hypotheses relating to the parameters of
taxation. Section 5 describes the empirical specification. Section 6 presents the
empirical findings and provides a sensitivity analysis. Section 7 concludes with
a summary of the most important findings.
2 Previous research
Under which conditions and to which extent corporate taxes influence a firm’s
location and production decisions is lively debated, not only among policy mak-
ers but also among researchers (see Hines, 1997, 1999; Gresik, 2001; Devereux,
2006, for comprehensive surveys). If firms cannot arbitrarily shift their profits
abroad, taxes reduce their after-tax profits and this, in turn, affects both the
location and the volume of FDI. Then, a high tax burden in a host country
represents an impediment to its inbound FDI, even if its effect is partly offset
to the extent that governments use tax revenues to reduce investment costs.
In fact, this reasoning may explain why several industrialized countries have
recently reduced their corporate tax rates.1 For instance, in the Western Euro-
1Within the OECD, the statutory corporate tax rate (excluding local corporate income
taxes) fell by 15 percent between 2000 and 2005, where the strongest reductions took place in
Austria (2005, from 34 to 25 percent), Belgium (2005, from 39 to 33 percent), Canada (several
reductions 2000-2005, from 28 to 21 percent), Germany (2001, from 40 to 25 percent), Iceland
(2002, from 30 to 18 percent), Ireland (several reductions 2000-2005, from 24 to 12.5 percent)
and Luxembourg (2002, from 30 to 22 percent). Among the Eastern European members, the
lowest levels of corporate tax rates amount to 16 percent (Hungary, since 2004) and 19 percent
(Poland and Slovak Republic, since 2004), respectively.