Working Paper: NBER ID: w2962
Authors: Patric H. Hendershott; Yun Hi Won
Abstract: Model simulations are run to obtain a range of realistic estimates of the long-run revenue impact of a capital-gains tax-rate cut to a maximum of 15 percent. The basic vehicle for the simulations is a slightly modified version of the Galper-Lucice-Toder (GLT) general equilibrium model. The key behavioral assumptions affecting the estimates are: (1) the portfolio and tangible capital reallocations implicit in the structure of the GLT model, (2) corporate payouts responses based on recent empirical estimates, and (3) illustrative noncorporate recharacterizations of regular income as capital gains. The essential message of this paper is that the strong emphasis in the literation on the realization response to a capital gains tax rate cut has been appropriate. The payout/recharacterization and portfolio redistribution/reallocation effects do not appear to be large. Moreover, the portfolio responses, within the context of the GLT model, act to raise tax revenues (substitution of taxable business capital for tax free household and state and local capital), not lower them as has been conjectured. Thus these responses offset the payout/recharacterization effects, leaving the realization response as basically the total response. Future research could, of course, modify this finding.
Keywords: capital gains tax; federal tax revenues; capital allocation
JEL Codes: H24; H25
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Capital gains tax rate cut (H24) | increase in realizations (F69) |
| increase in realizations (F69) | federal tax revenues (H29) |
| increase in realizations (F69) | reduction in corporate income paid as dividends (G35) |
| reduction in corporate income paid as dividends (G35) | federal tax revenues (H29) |
| portfolio reallocations (G11) | federal tax revenues (H29) |
| tangible capital stock adjustments (E22) | federal tax revenues (H29) |