« Back to Results

Estimation and Interpretation of Tax Distortions

Paper Session

Saturday, Jan. 6, 2018 10:15 AM - 12:15 PM

Pennsylvania Convention Center, 106-B
Hosted By: Econometric Society
  • Chair: Nathan Seegert, University of Utah

Better Bunching, Nicer Notching

Andrew Harrison McCallum
,
Federal Reserve Board
Nathan Seegert
,
University of Utah
Marinho Bertanha
,
University of Notre Dame

Abstract

Bunching and notching estimation methods pioneered by Saez (2010) and Kleven and Waseem (2013) use mass points in the distributions of income or profits that are caused by policies to estimate key parameters in public finance. The goal of this paper is to build on and enhance these estimators by combining advances in public finance, labor economics, and econometrics. The result is an updated method that retains the insights and advantages of the original bunching and notching estimators while incorporating covariates. In particular, we place our estimator within the econometrics literature by showing that bunching and notching are censoring models. Once our estimator is recast in this light, it is possible to leverage the significant econometric developments since Tobin (1958) and to place bunching and notching on a rigorous statistical foundation. We compare the updated and standard methods using Monte Carlo simulations to illustrate the relative performance of recovering parameter values. Finally, we apply our method in the context of the earned income tax credit to show it leads to quantitatively different estimates of the compensated elasticity of reported income with respect to (one minus) the marginal tax rate.

Bunching versus Diff-in-Diff

Miguel Almunia
,
University of Warwick
Michael Carlos Best
,
Stanford University

Abstract

How can we reconcile the wide differences between difference-in-difference estimates and bunching estimates of taxable income elasticities? We provide a simple framework showing how the dynamics of bunching estimates and difference-in-difference estimates around the time of tax reforms can shed light on the role of optimization frictions in determining taxable income responses. We implement our method on U.K. data and a reform to the top income tax brackets.

Malas Notches

Ben Lockwood
,
University of Warwick

Abstract

This paper shows that the sufficient statistic approach to the welfare properties of income (and other) taxes does not extend to tax systems with notches, because with notches, changes in bunching induced by changes in tax rates have a first-order effect on tax revenues. In an income tax setting, we show that the marginal excess burden (MEB) and the welfare-maximizing top rate of tax are given by the relevant formulae for a proportional tax as in Feldstein (1999) plus a correction factor. The Feldstein formulae always underestimate the MEB and overestimate the revenue and welfare-maximizing rate of tax. Quantitatively, these mis-estimates can be very large; the MEB can be underestimated by an order of magnitude. An application to VAT is discussed; with a calibration to UK data, the MEB of the VAT is roughly three times what is would be if VAT was simply a proportional tax.

Tax Advantages and Imperfect Competition in Auctions for Municipal Bonds

Daniel Garrett
,
Duke University
Andrey Ordin
,
Duke University
James W. Roberts
,
Duke University
Juan Carlos Suarez Serrato
,
Duke University

Abstract

We show that the effect of tax advantages of municipal bonds on the market structure of municipal
bond auctions is a crucial determinant of state and local governments’ borrowing costs. Reduced-form estimates show that increasing the tax advantage by 3-pp. lowers mean borrowing costs by 9-10%, consistent with a greater-than-unity passthrough elasticity. Non-parametric evidence shows that strategic participation and bidding in imperfectly-competitive auctions generates this greater-than-unity passthrough. Using a structural auction model to evaluate the efficiency of Obama and Trump administration proposals, we find that the reduction in municipal borrowing costs is 2.8-times the revenue cost of the tax advantage
Discussant(s)
Daniel Wilson
,
Federal Reserve Bank of San Francisco
Caroline Weber
,
University of Oregon
Joel Slemrod
,
University of Michigan
John N. Friedman
,
Brown University
JEL Classifications
  • H20 - General
  • H21 - Efficiency; Optimal Taxation