ROW: Distributional Effects of Distortionary Taxation

该项目计划研究扭曲税收政策的分配效应。在大多数对税制改革的分析中，税收政策如何影响具有不同偏好、天赋或投资机会的代理人的问题没有得到解决。在动态经济的情况下，这种遗漏的主要原因是，这项任务在计算上是令人敬畏的。然而，初步迹象表明，流体力学和结构力学中用于求解偏微分方程组的数值方法可能是适用的。这项研究考察了它们在具有异质代理人和扭曲税收的模型中的应用。该项目需要几项规划活动。首先是在已知解决方案的情况下测试这些方法。这类问题通常涉及一个没有扭曲的经济，由相同的代理人居住。第二是在具有相同代理人和扭曲税收的模型上测试这些方法。对于一些扭曲的经济模型，存在参数或变量的转换，这使得它类似于一个没有扭曲的经济。对于这两组测试用例，将使用替代的数值方法来验证结果。初步结果表明，有限差分技术表现良好，可能足以有效地进行最大似然估计。最后，还将探讨可供选择的有限元和变分方法。

This project plans to study the distributional effects of distortionary tax policy. In most analyses of tax reform, the question of how tax policy affects agents with different preferences, endowments or investment opportunities is not addressed. In the case of dynamic economies, the main reason for this omission is that the task is computationally formidable. However, preliminary indications show that numerical methods used in fluid dynamics and structural mechanics for solving partial differential equations may be applicable. This research investigates their application to a model with heterogeneous agents and distortionary taxation. The project requires several planning activities. The first is to test the methods on cases for which there is a known solution. Such problems usually involve a distortion-free economy inhabited by identical agents. The second is to test the methods on models with identical agents and distortionary taxes. For some models of a distorted economy, there exists a transformation of parameters or variables which makes it resemble a distortion-free economy. For both sets of test cases, alternative numerical methods will be used to validate the results. Preliminary results show that finite difference techniques perform well and may be sufficiently efficient for maximum likelihood estimation. Finally, the alternative finite-element and variational methods will also be explored.