Federal Tax-Transfer Policy and Intergovernmental Pre-Commitment

Table 1: Decentralized leadership: comparative static analysis.

subject to kⁱ = kⁱ(tⁱ, t^j), sⁱ = sⁱ(tⁱ, t^j) and τ = τ(tⁱ, t^j). The first-order condition is

V_tⁱi +Vτⁱτti +Vgⁱ(tⁱk_tⁱi +kⁱ+sⁱ_ti) =0.                             (21)

Besides the direct effects of state i’s tax policy on its constituent’s private and public consump-
tion, state i’s taxing incentives are augmented by two strategic effects. State i also takes into
account how the induced change in the federal policy variables affects utility derived from pri-
vate consumption, V_τⁱτ_ti > 0, and utility derived from local public good provision, V_gⁱsⁱ_ti . If
V_τⁱτ_ti + V_gⁱsⁱ_ti 6= 0, capital taxes in the decentralized leadership game differ from capital taxes
chosen under a Nash conjecture. Denoting (t^d, τ^d, s^d) as the prevailing policy choices Proposi-
tion 3 relates the strategic effects to the sign of α.

Proposition 3: If α > (<) 0, the tax mix under decentralized leadership is less (more) ef-
ficient relative to the tax mix under Nash behavior, i.e. t^d < (>) t^N and τ^d > (<) τ^N . If
α = 0, the tax mix under decentralized leadership and Nash behavior is identical, i.e. t^d = t^N
and τ^d = τ^N.

The proof is relegated to the appendix. The term α is decisive in signing the tax differential
t^d - t^N which is inversely related to the tax differential τ^d - τ^N . In a symmetric equilibrium

19

<<   17   18   19   20   21   22   23   >>

More intriguing information