Factor income distribution and growth regimes in Latin America, 1950–2012

• Author: Germán ALARCO
• Date: 01 March 2016
• DOI: http://doi.org/10.1111/ilr.12006
• Published date: 01 March 2016

Copyright © The author 2016

Journal compilation and translation © International Labour Organization 2016

International Labour Review, Vol. 155 (2016), No. 1

Factor income distribution and growth

regimes in Latin America, 1950–2012

Germán ALARCO*

Abstract. After a comprehensive review of the literature on economic growth re-

gimes, the author develops an econometric model to determine the type of growth

regime prevailing in 16 Latin American countries over the period 1950–2012 , using

statistical data on the respective wage share and prot share of GDP. After ana-

lysing the evolution of factor income distribution in relation to economic growth

during the period in question, the author concludes that most Latin American

economies have a wage-led growth regime, and that redistribution policies targeted

at wages are therefore conducive to economic growth.

The issue of income distribution has recently returned to the top of the

global agenda following the work of Piketty (2014), who provides sys-

tematic long-term statistical data on the concentration of wealth and income

growth in the major world economies. While it was not his explicit intention,

Piketty has opened up debate on the third of the three basic economic ques-

tions – what to produce, how to produce, and for whom to produce? – in the

classical and post-Keynesian tradition. For Piketty, high levels of wealth and

income inequality represent a threat to democracy; he does not, however, ex-

plore the ways in which this inequality also contributes to crises, economic

downturns and social disorder.

In recent decades, the issue of income distribution has also received re-

newed attention on the research agenda in Latin America. Emphasis is placed

on personal income distribution, however, rather than on factor or functional

income distribution, on which less work has been carried out.1 The latter

* Research professor, Universidad del Pacíco (Peru), Postgraduate School, email:

g.alarcotosoni@up.edu.pe. The author gratefully acknowledges the help received from César

Castillo and Kevin Gershy-Damet, research project assistants, and Favio Leiva, who contributed to

an earlier study on the same subject. He also thanks Patricia del Hierro Carrillo for her comments.

Responsibility for opinions expressed in signed articles rests solely with their authors, and

publication does not constitute an endorsement by the ILO.

1

Work that has been carried out includes Abeles, Amarante and Vega (2014), Alarco (2014a),

ILO (2012), ECLAC and ILO (2012), Neira Barría (2010), Lindenboim (2008), Frankema (2009),

Bértola et al. (2008) and Fitzgerald (2009).

International Labour Review74

approach poses a number of problems, such as the fact that in most Latin

American economies mixed income or self-employed income is included in

the gross operating surplus, resulting in corresponding adjustments and esti-

mates being made in some studies.

Analysing personal income distribution is not straightforward either,

however, since the household surveys on which such studies are based under-

estimate the highest income strata, and there are mismatches when these

results are extrapolated from national accounts. In addition, statistics on the

subject are hard to come by in Latin American countries; in most of them, ac-

cess to data based on income tax returns is not possible, and the absence of

inheritance tax makes it impossible to collect statistical data on the evolution

of wealth accumulation of the kind compiled by Piketty (2014).

Since Kalecki (1954), factor income distribution has been a central ele-

ment in explaining the level and evolution of GDP. Post-Keynesians have ex-

plored this relationship in depth, including researchers investigating growth

regimes, who analyse the various ways in which the wage share of GDP inu-

ences private consumption, private investment, exports and productivity, seek-

ing to establish whether an increased wage share contributes to an increase in

economic activity, or not. If so, we could say that the growth regime is wage-

led; if not, that the growth regime is prot-led.

This article aims to review the evolution of the wage share and prot

share of GDP (specically, the share of wages and salaries and the gross op-

erating surplus, respectively) in Latin America, for the period 1950–2012 , and

to determine whether the 16 economies analysed were wage-led or prot-led.

This article builds on an earlier study (Alarco, 2014a) on the wage share. The

main questions addressed are: What does the literature on the relationship be-

tween income distribution and economic growth regimes tell us? What is the

distribution pattern of the wage share and prot share in Latin America? Are

there common distribution patterns for the entire region, or are some features

specic to particular groups of countries? How has the wage share evolved

compared to economic growth in the long run? In the different economies of

Latin America, is the growth regime wage-led or prot-led?

The remainder of the article is organized into ve sections. The rst sec-

tion reviews the literature on the relationship between income distribution

and growth regimes. The second describes the model used to determine the

type of economic growth regime, and the data sources used. The third section

describes the basic statistical data on the evolution of the wage share and

prot share, common distribution patterns, and the correlation between wage

share and economic growth in Latin America as a whole. In the fourth sec-

tion the selected model is applied and the results presented, thereby identify-

ing what economic growth regime prevailed during the period under review

in the countries studied. The last section presents a number of conclusions and

policy recommendations.

With regard to the article’s scope and limitations, in line with our macro-

economic perspective we omit structural and socio-political factors specic