> In contrast to Chamley-Judd, the optimal tax on capital is positive in our model because we have finite long run elasticities of inheritance to tax rate

https://www.nber.org/papers/w17989.pdf