a unit of tax revenue goes up, so financing the public sector wage bill becomes
costlier in economic and political terms. The vote-maximizing political strategy
will then involve lower rents to public sector workers and fewer public sector jobs
(and hence less public service provision).
A main point in this explanation of the impact of tax competition on the
political equilibrium is that the public sector workers earning rents are part of the
voting population, so a cut in rents comes at a cost to vote-maximizing politicians.
Obviously, this political cost arises from the fact that the fall in rents induced
by tax competition reduces the utility of public sector workers. By recognizing
this, our framework allows us to identify an optimal degree of tax competition,
accounting for the welfare of all citizens. The standard Leviathan models of tax
competition are unsuited for this purpose because they postulate that rents do not
generate welfare for any citizen, and because they do not account for the fact that
tax competition - by changing the size of the public sector - will also change the
composition of the voting population, thereby affecting the political equilibrium
and the distribution of welfare.
Qualitatively, tax competition affects social welfare through three different
channels in our model: 1) By driving a wedge between the marginal rate of sub-
stitution and the marginal rate of transformation between public and private
goods, it tends to reduce aggregate welfare. This is the welfare-reducing effect of
tax competition emphasized in the literature assuming that policy is made by a
benevolent social planner. 2) By reducing rents to public sector workers, tax com-
petition equalizes the marginal rate of substitution between public and private
goods across private and public sector workers. Ceteris paribus, this results in
an outward shift of the utility possibility frontier which tends to increase social
welfare. 3) By curbing rents to public sector workers, tax competition also equal-
izes the marginal utility of income for all citizens. In a society concerned about
equality, this likewise tends to increase social welfare. By including the mechan-
isms 2) and 3), our analysis accounts for both of the standard criticisms against
rent-seeking, i.e., the objections that it inefficient as well as unfair.
We first consider the case where governments non-cooperatively choose the
level of capital taxation, public sector rents and employment. We show that
an increase in tax base mobility will initially tend to be welfare-increasing, but
beyond a certain point which depends inter alia on the political strength of the
public sector lobby, a further increase in tax base mobility will reduce welfare