Tax impact on multi-stage mean-variance portfolio allocation

We investigate the sensitivity to tax change of multi-stage portfolio allocation over a discrete time investment horizon. Special taxation rules within wrappers grouped a number of risky assets are integrated with multi-stage linear or quadratic stochastic programming in the mean-variance framework. The uncertainty on the returns of assets is specified as a scenario tree generated by a simulation-based approach. We adjust different values on capital gains tax under different asset bounds and risk levels. The tax impact in the yearly reallocation of the investments for a typical case with an annual fixed withdrawal that utilizes completely the option of taper relief is also explored. Our computational results show that taxes, combined with other effects such as risk and investment upper bounds, have a significant performance impact on portfolio allocation as well as diversification over wrappers. Yet, investment strategies can be made robust to changes in taxation.