How are optimal taxes affected by superstar phenomena? To answer this question, we extend the Mirrlees model to incorporate an assignment problem in the labor market that generates superstar effects. Perhaps surprisingly, rather than providing a rationale for higher taxes, we show that superstar effects provide a force for lower marginal taxes conditional on the observed distribution of earnings. Superstar effects make the earnings schedule convex, which increases the responsiveness of individual earnings to tax changes. We show that various common elasticity measures must be adjusted upward in optimal tax formulas. Finally, we study a comparative static that does not keep the observed earnings distribution fixed: when superstar technologies are introduced, inequality increases but we obtain a neutrality result, finding optimal taxes unaltered.

This paper studies optimal dynamic tax policy under the threat of political reform. A policy will be reformed ex post if a large enough coalition of citizens supports reform; thus, sustainable policies are those that will continue to attract enough political support in the future. We find that optimal marginal capital taxes are either progressive or U-shaped, so that savings are subsidized for the poor and/or the middle class but are taxed for the rich. U-shaped capital taxes always emerge when individuals' political behavior is purely determined by economic motives.

In the presence of a time-inconsistency problem with agency contracts, we show that competitive markets can implement allocations that Pareto-dominate those achieved by a benevolent government, and they induce more effort. We analyze a model with moral hazard and a two-sided lack of commitment. After agents have chosen their work, firms can modify contracts and agents can switch firms. If the ex post market outcome satisfies a weak notion of competitiveness and sufficiently separates individuals, it is Pareto superior to a government’s allocation with a complete breakdown of incentives. Moreover, competitive markets without commitment implement more effort in equilibrium under general conditions.

We study competitive equilibria in a signalling economy with heterogeneously informed buyers. In terms of the classic Spence (1973, The Quarterly Journal of Economics, 87, 355—374) model of job market signalling, firms have access to direct but imperfect information about worker types, in addition to observing their education. Firms can be ranked according to the quality of their information, i.e., their expertise. In equilibrium, some high-type workers forgo signalling and are hired by better informed firms, which make positive profits. Workers’ education decisions and firms’ use of their expertise are strategic complements, allowing for multiple equilibria that can be Pareto ranked. We characterize wage dispersion and the extent of signalling as a function of the distribution of expertise among firms. Our model can also be applied to a variety of other signalling problems, including securitization, corporate financial structure, insurance markets, or dividend policy.

We develop a framework for optimal taxation when agents can earn their income both in traditional activities, where private and social products coincide, and in rent-seeking activities, where private returns exceed social returns either because they involve the capture of pre-existing rents or because they reduce the returns to traditional work. We characterize Pareto optimal income taxes that do not condition on how much of an individual's income is earned in each of the two activities. These optimal taxes feature an externality-corrective term, the magnitude of which depends both on the Pigouvian correction that would obtain if rent-seeking incomes could be perfectly targeted and on the relative impact of rent-seeking externalities on the private returns to traditional and to rent-seeking activities. If rent-seeking externalities primarily affect other rent-seekers, for example, the optimal correction lies strictly below the Pigouvian correction. A calibrated model indicates that the gap between the Pigouvian and optimal correction can be quantitatively important. Our results thus point to a hefty informational requirement for correcting rent-seeking externalities through the income tax code.

We consider optimal redistribution in a model where individuals can self-select into one of several possible sectors based on heterogeneity in a multidimensional skill vector. We first show that when the government does not observe the sectoral choice or underlying skills of its citizens, the constrained Pareto frontier can be implemented with a single nonlinear income tax. We then characterize this optimal tax schedule. If sectoral inputs are complements, a many-sector model with self-selection leads to optimal income taxes that are less progressive than the corresponding taxes in a standard single-sector model under natural conditions. However, they are more progressive than in canonical multisector economies with discrete types and without occupational choice or overlapping sectoral wage distributions.