By Michael Kuklik and Nikita Céspedes
The optimal capital income tax rate is 36 percent as reported by Conesa, Kitao, and Krueger (2009). This result is mainly driven by the market incompleteness as well as the endogenous labor supply in a life-cycle framework. We show that this model fails to account for the basic life-cycle features of the labor supply observed in the U.S. data. In this paper, we introduce into this model non-linear wages and inter-vivos transfers into this model in order to account for the life-cycle features of labor supply. The former makes hours of work highly persistent and helps to account for labor choices at the extensive margin over the life cycle. The latter allows us to account for labor choices early in life. The suggested model delivers an optimal capital income tax rate of 7.4 percent, which is significantly lower than what Conesa, Kitao, and Krueger (2009) found.
One more paper in an everexpanding literature on optimal capital income taxation. And once more, it shows how dramatically sensitive results are to modeling assumptions. Is this literature ever going to come to a conclusion? On one hand, this challenge makes it all the more exciting for the researcher, on the other hand, the lack of robustness of results is quite disheartening.