Top Incomes in Chile: A Historical Perspective on Income Inequality, 1964–2017

• Author: Claudia Sanhueza,Ignacio Flores,Ricardo Mayer,Jorge Atria
• DOI: http://doi.org/10.1111/roiw.12441
• Published date: 01 December 2020
• Date: 01 December 2020

© 2019 Internation al Association for Re search in Inco me and Wealth

850

TOP INCOMES IN CHILE: A HISTORICAL PERSPECTIVE ON INCOME

INEQUALITY, 1964–2017

by IgnacIo Flores*

INSEAD and World I nequality Labo ratory (WIL)

claudIa sanhueza

Center for Eco nomics and Socia l Policy and Faculty of Hu manities,Univers idad Mayor

Center for Soci al Conict and Co hesion Studies (COES), Chile

Jorge atrIa

Center for Eco nomics and Socia l Policy and Faculty of Hu manities,Univers idad Mayor

Center for Soci al Conict and Co hesion Studies (COES), Chile

AND

rIcardo Mayer

Economic C ommission for La tin America (ECLAC)

We present a novel series of Chilean top-income shares covering half a century, mainly based on income-

tax declarations and the National Accounts. Such a time frame of analysis is still rare in the literature of

developing countries. We distinguish between a fiscal-income series (1964–2017) and an adjusted series

(1990–2017). The former covers individual income, while the latter also includes corporate undistrib-

uted profits, which affects both levels and trends. The fiscal-income estimates start with low levels and a

decreasing trend over the 1960s. They then increase rapidly during the dictatorship years (1973–89). The

series ends with a high, yet slowly decreasing, concentration for most of the recent democratic period

(1990–2017). By contrast, the adjusted series has followed a U-shape since the return of democracy,

contradicting the established consensus on falling inequality over the period. Furthermore, Chile ranks

among the most unequal countries in both the OECD and Latin American countries over the period.

JEL Codes: D31, N36, H24

Keywords: Chile, historical trends, inequality, tax data, top incomes

Note: All files and codes that were necessary to produce the estimates and tables presented in this

paper are publicly available in Stata format at https ://github.com/ignac io-flore s/top-incom es-chile.git.

The authors are grateful to Facundo Alvaredo, Thomas Blanchet, Mauricio De Rosa, Amory Gethin,

Pablo Gutiérrez, Thanasak Jenmana, Marc Morgan, Francisco Saffie, and two anonymous referees for

helpful suggestions and comments. I.F. gratefully acknowledges funding from the Ford Foundation.

J.A. thanks CONICYT, FONDECYT No. 3160705, and FONDAP No. 15130009 and C.S. thanks

CONICYT, FONDECYT No. 11110470, and FONDAP No. 15130009, for providing the funding for

this research.

*Correspondence to: Ignacio Flores, Postdoctoral Researcher at INSEAD—Wealth Inequality

Center, Bvd. de Constance, 77300 Fontainebleau, France, Of. EW 2.12; Research Fellow at World

Inequality Laboratory, PSE 48 Bvd. Jourdan, 75014 Paris, France, Of. 5-01 (ignacio.flores@insead.

edu).

Review of Inc ome and Wealth

Series 66 , Number 4, Decemb er 2020

DOI : 10.1111 /roi w.124 41

bs_bs_banner

Review of Income and Wealth, Series 66, Number 4, December 2020

851

© 2019 Internation al Association for Re search in Inco me and Wealth

1. IntroductIon

Following seminal contributions by Piketty (2001) and Piketty and Saez

(2003), extensive progress has been made over the past two decades in the field of

economic inequality. Studies on more than 40 countries have used tax data to

explore income concentration within the richest fractiles of the population.1 These

works have revealed sections of the distribution that were previously invisible to

the eyes of researchers, thus allowing the examination of a larger part of it and

extending farther back in time than any survey statistic. Indeed, the true value of

tax statistics is to focus on small groups of people in whom substantial parts of

total income are concentrated, the evolution of which is likely to influence overall

inequality trends (Alvaredo etal., 2013).

However, to date there is still scant evidence on the long-term evolution of

top-income shares in the developing world. This paper contributes to filling this

gap by making use of a series of Chilean tax statistics that extends over more than

50 years. Chile is an interesting case for various reasons. Although ranked among

the most unequal OECD countries (OECD, 2015), Chile has been considered one

of the stronger states in Latin America in terms of state capacity, corruption levels,

and the effectiveness of tax policy. Yet the country still has a relatively low level

of redistribution, and fiscal policy has a limited capacity to reduce extremely high

market inequalities (OECD, 2015).

This is certainly not the first time that income-tax declarations have served as

a base for the study of Chilean inequality. Previous works by López et al. (2013)

and Fairfield and Jorratt De Luis (2016) provide solid estimates on top-income

shares in the country, yet only for a few recent years.2 In this paper, we use these

studies—especially Fairfield and Jorratt De Luis (2016), due to the data quality—

as valuable references to adjust and compare the level of our own estimates, which

are mostly focused on describing long-run trends. Furthermore, both studies have

also highlighted the local relevance of studying undistributed profits, which likely

have a biasing impact on inequality estimates via incentives to retain profits inside

corporations. Hence, in this study, we distinguish between our fiscal-income series,

which strictly covers individual income for the period from 1964 to 2017, and the

adjusted series, which includes the imputation of undistributed profits for a shorter

period from 1990 to 2017.

Our findings indicate that income concentration remains high in both series

throughout the whole period under study. In a broad perspective, the fiscal-income

series starts with relatively low levels—with top 1 percent income shares close to

13 percent on average—and a decreasing concentration during the 1960s and the

beginning of the 1970s. The trend is sharply interrupted in 1974, when rapidly

increasing concentration begins and holds at least for the first half of Pinochet’s

dictatorship, reaching a top 1 percent close to 17 percent in 1981. Due to miss-

ing information, the series is interrupted for the rest of the regime. The return to

democracy takes place in a context of even higher inequality, with a top 1 percent

1See the World Inequality Database at http://www.WID.world .

2López et al. (2013) covers the period from 2004 to 2010, López et al. (2016) the period from 2004

to 2013, and Fairfield and Jorratt De Luis (2016) only have data for 2005 and 2009.