We analyze the optimal nonlinear income tax schedule when taxpayers earn multiple incomes and differ along many unobserved dimensions. We derive the necessary conditions for the government’s optimum using both a tax perturbation and a mechanism design approach, and show that both methods...

We analyze the optimal nonlinear income tax schedule when taxpayers earn multiple incomes and differ along many unobserved dimensions. We derive the necessary conditions for the government's optimum using both a tax perturbation and a mechanism design approach, and show that both methods produce...

There is increasing empirical evidence that people systematically differ in their rates of return on capital. We derive optimal non-linear taxes on labor and capital income in the presence of such return heterogeneity. We allow for two distinct reasons why returns are heterogeneous: because...

We study the optimality of taxing capital income according to a Rate-of-Return Allowance proposed by the Mirrlees Review. In a mean-variance framework the optimal tax on risk-free returns is zero with constant returns to scale in private investment, but positive with decreasing returns to scale,...

There is increasing empirical evidence that people systematically differ in their rates of return on capital. We derive optimal non-linear taxes on labor and capital income in the presence of such return heterogeneity. We allow for two distinct reasons why returns are heterogeneous: because...

This paper studies the design of tax systems that implement a planner's second-best allocation in a market economy. An example shows that the widely used Mirrleesian (1976) tax system cannot implement all incentive-compatible allocations. Hammond's (1979) principle of taxation proves that any...

We characterize the second-best allocation in a Mirrleesian optimal tax model where agents differ in multiple dimensions and the planner can tax multiple goods non-linearly. We develop a new method that allows us to solve the partial differential equations that describe the optimum regardless of...

This paper develops a machine-learning method that allows researchers to estimate heterogeneous treatment effects with panel data in a setting with many covariates. Our method, which we name the dynamic causal forest (DCF) method, extends the causal-forest method of Wager and Athey (2018) by...

This paper develops a machine-learning method that allows researchers to estimate heterogeneous treatment effects with panel data in a setting with many covariates. Our method, which we name the dynamic causal forest (DCF) method, extends the causal-forest method of Wager and Athey (2018) by...

Understanding the effectiveness of social distancing on the spread of COVID-19 is crucial to justify economically costly social distancing measures. We present a case study focusing on the three Scandinavian countries. Whereas Denmark and Norway imposed relatively strict measures, Sweden follows...