Optimal Income Taxation with Multidimensional Taxpayer Types
Beginning with Mirrlees, the optimal taxation literature has generally focused on economies where individuals are differentiated by only their productivity. In this pa- per we examine models with discrete types where individuals are differentiated by two or more characteristics. For example, we examine the case where individuals also have different tastes for labor supply. We find that the extra dimensionality produces substantively different results, such as negative marginal tax rates for some high pro- ductivity taxpayers. This is a difficult nonlinear optimization problem because quite often, the linear independence constraint qualification does not hold at a solution. To robustly solve these degenerate nonlinear programs, we use the solver SNOPT, an SQP method with an elastic-mode. Our computational framework proposed in this paper can also be applied to solve problems that arise in multidimensional nonlinear pricing and mechanism design
When requesting a correction, please mention this item's handle: RePEc:sce:scecfa:471. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Christopher F. Baum)
If references are entirely missing, you can add them using this form.