more—because of a stronger expenditure switching effect—in the revenue-neutral policy considered here.
Our exercise is similar in spirit to the one carried out by Mendoza and Tesar (1998) in an open economy RBC model with capital. The results are not directly comparable, since they look at the implications of a much more radical policy than ours, in which the US federal income tax is completely eliminated and replaced with consumption taxes. It is interesting, however, to notice that in their case—unlike in our model—this would imply a positive effect on domestic output and consumption both in the long run and along the transition path, while the impact on foreign output is positive in the short run but negative in the long run. While it is difficult to draw strong conclusions from the comparison, one possible explanation of the different results that we get is again that in our model, due to the presence of nominal rigidities, the expenditure switching effect implies that a stronger part of the benefit of income tax reductions accrues to the foreign country. This can, at least partially, explain why the domestic effect of a revenue neutral policy is expansionary in their model but contractionary in ours.
VII. SENSITIVITY ANALYSIS
In this section we present the results of some sensitivity analysis. In particular, we look at how the degree of self-financing changes when we change the degree of nominal rigidity γ and the labor disutility parameter ν . Overall, our sensitivity analysis confirms the
robustness of the results that we have discussed in sections IV and V.
Income Tax Cut
Table 6 shows that the results on the degree of self-financing in real terms are robust to changes in the degree of nominal rigidity in the economy. In particular, the degree of self financing of real revenue collection stabilizes around 17 percent in the new steady state regardless of the value of γ . In the case of γ = 0 the nominal degree of self-financing is
zero, since in a perfectly competitive labor market with no rigidities and a unitary disutility parameter the nominal wage response is a mirror image of the increase in the labor supply
ˆ stemming from the tax cut ( wˆ = −l ).14
In Table 7 we show how our benchmark results change when we vary the labor disutility parameter along the range estimated for it by Rotemberg and Woodford (1997). Not surprisingly, Table 7 shows that a decrease in the disutility parameter of supplying labor increases the degree of self-financing of labor income tax cuts (and vice versa). Even when we set ν = 0.47 (at the bottom of the range estimated by Rotemberg and Woodford (1997)), however, our estimated degree of self-financing remains close to 20 percent.
This implies that the general equilibrium and the partial equilibrium effects are the same in eq. (26) because
ˆ wˆ and l compensate each other in the numerator.