Page 1 of 37

Rising Earnings Inequality and Optimal Income Tax And Social

Security Policies✩

Pavel Brendler

University of Bonn, Institute for macroeconomics, Adenaueralle 24-42, 53113 Bonn, Germany.

Abstract

How did the US government preferences over income redistribution across generations and

within generations change during 1980–2010? Using a rich quantitative model in which a

Ramsey government chooses income taxation and Social Security, I decompose the total

change in the actual policies into the impact of new economic and demographic conditions

and government preferences. I find that the US government preferences have shifted toward

more educated and older households since the 1980s. Preferences over income redistribution

within and across generations interact and, therefore, must be analyzed jointly.

Keywords: Optimal Tax, Progressive Taxation, Social Security, Earnings inequality,

Aging Population

JEL: D3, E6, H2, H3

✩I am particularly thankful to Yongsung Chang and an anonymous referee for very useful comments and

suggestions. I benefited a lot from insightful discussions with Pawel Doligalski, Emanuel Hansen, Thomas

Hintermaier, Byeongju Jeong, Moritz Kuhn, and Benjamin Larin. I also received very valuable feedback from

Arpad Abraham, Marco Bassetto, Gregor Boehl, Klaus Gruendler, Fatih Guvenen, Jonathan Heathcote, Ur- ban Jermann, Joachim Jungherr, Sagiri Kitao, Winfried Koeniger, Tim Krieger, Dirk Krueger, Patrick Mac- namara, Alisdair McKay, Christian Moser, Roberto Pancrazi, Facundo Piguillem, Dominik Sachs, Almuth

Scholl, Ctirad Slavik, Kjetil Storesletten, Arnau Valladares-Esteban, Chunzan Wu as well as the seminar

participants at several seminars and conferences. I thank Oliver Freyermuth at the Physics Department of

the University of Bonn for his excellent IT support. As always, all mistakes are mine.

Email address: pavel.brendler@uni-bonn.de (Pavel Brendler)

Preprint submitted to Elsevier September 29, 2022

Page 2 of 37

1. Introduction

The recent macroeconomic literature has extensively studied the optimal income tax- and-transfer system, largely ignoring the role played by Social Security, a publicly provided

pension system in the US. Since the two programs redistribute incomes both across genera- tions and within generations, their joint analysis allows to study how government preferences

over income redistribution have changed over time, which is the paper’s main focus.

One of the paper’s key innovations is to relax the commonly used assumption in the

optimal income taxation literature that the Ramsey government maximizes the welfare of

newborns. Instead, the policymaker cares about all agents who are alive at the time when the

policy is implemented. This departure is crucial because newborn workers in the calibrated

model economy prefer to shut down the public pension system altogether, regardless of their

characteristics, so the model is unable to rationalize why Social Security and the income

tax-and-transfer program coexist. The second significant deviation from the existing studies

is the introduction of age- and education-specific Pareto weights, which allow the model to

account for the actual sizes of both programs in the data. By quantifying the change in the

joint distribution of Pareto weights by age and education during 1980–2010, this paper is

able to analyze the evolution of the US government preferences over income redistribution

across generations and within generations, which is the paper’s main contribution.

The government has access to two policy instruments in the model: the income tax

progressivity and the average replacement rate in the pension system. The government sets

both policies jointly, internalizing how the policy will affect all currently alive agents along

the transition. Workers accumulate skills and make retirement decisions, so both income

taxes and payroll contributions exert a distortionary effect on their decisions. In line with

the literature, the policy reform is unanticipated and permanent.

Through the lens of the model, two distinct forces are responsible for the change in

the observed policies during 1980–2010. The government might have found it necessary to

adjust both policies because it faced new economic and demographic conditions in the 2010s

2