capital taxes rise which inclines the federal government to rely less on distortionary labor taxes.
The net effect on total tax revenues turns out to be positive, which leaves public good provision
at a more efficient level. The rationale is that local government do not only compete for mobile
capital, but also for federal transfers. The federal government allocates transfers dependent on
the inequality of public consumption across states. If a state collects more tax revenues relative
to the neighboring state (due to higher taxing effort), it forfeits transfers which undermines
taxing incentives. With capital mobility the neighboring state also collects more capital tax rev-
enues since capital moves to states which offer a more attractive tax policy. If capital mobility
is sufficiently high, the neighboring state enjoys an increase in tax revenues which outweighs
the increase in the tax-raising state. In this situation federal transfers are re-shuffled to the
tax-raising state - a positive transfer effect which counteracts the traditional tax competition
effect.
The outline of the paper is as follows. The related literature is reviewed in Section 2.
Section 3 introduces the model. The policy outcome in the case that both levels of government
decide simultaneously on policy instruments is presented in Section 4. Section 5 examines the
interaction between the federal and state level if the federal government pre-commits, while
Section 6 discusses the efficiency implications if state governments pre-commit. Section 7 offers
some conclusions.
2 Related Literature
An analytical treatment of federal-local policy interaction is provided in Boadway and Keen
(1996), Boadway et al., (1998), Keen and Kotsogiannis (2002), and Boadway and Tremblay
(2006). In contrast to the fiscal arrangement considered in this paper they assume an overlap-