This paper extends the Mirrleesian non-linear income taxation model, of the modern form as expressed by Diamond (1997) and Saez (2001), to the case where people also care about their ordinal income rank in society. It is concluded that preferences for rank induce externalities that tend to increase the optimal marginal tax rate, and that these effects may be substantial. This is despite the fact that the ordinal income rank is unaffected in the model for all individuals who work, implying that the modified taxes due to rank positionality are not due to governmental incentives to change the ordinal rank. While this is the first Mirrleesian model with concerns for rank, the results are in contrast to some related earlier results.

We establish foundational results in dynamic consumption taxation. Network goods such as cellphones are an increasingly important part of the economy and require a different fiscal approach than non-network goods. Their consumption profile is dynamic, an aspect under-studied by the existing literature. Due to network effects, market growth today can increase willingness-to-pay tomorrow. This intertemporal externality changes the optimal tax rate through time. Optimality may require subsidization in early periods when demand is low followed by high subsequent taxes. These results also generalize to a wider class of goods with subtle network effects, including indoor dining during a pandemic.

This paper develops a comprehensive framework for analyzing the revenue, efficiency and social welfare implications of taxing top incomes. It generalizes the Saez (2001) formula for the optimal top tax rate by deriving analytical expressions for the Laffer curve and the excess burden. Applied to the 2021 U.S. top federal tax bracket, assuming a taxable income elasticity of 0.25, the study finds an excess burden of $101 billion and a maximum potential revenue increase of $111 billion. In contrast, other English-speaking countries and Germany are positioned closer to their Laffer curve peaks, whereas the Nordic countries studied are on the downward-sloping part of the Laffer curve. Additionally, the paper endogenizes the marginal social welfare weight on high-income earners and, following an inverse optimal taxation approach, concludes that in none of the studied countries does the observed top marginal tax rate appear consistent with a conventional welfarist social welfare function.

Measuring the welfare consequences of tax change is critical for policy evaluation. Tax schemes are progressive and piecewise functions, and agents decide on different occupations with different evasion facilities. This paper studies the measure of welfare effects of tax change incorporating those elements. We model an economy with those elements, where self-employed bunch at the income bracket threshold and declare less income to face a smaller marginal tax rate than they owe. Later, we characterize the welfare effect of tax change. This characterization depends on evasion, occupational decisions, and frictions based on the tax system. Those elements produce a divergence with the use of the elasticity of taxable income or total earned income. Also, we propose a metric to estimate the marginal change in welfare based on a trapezoid. Lastly, we estimate the measure using a tax reform in Chile, showing the relevance of incorporating occupational decisions and tax evasion.