Labour market turnover in Latin America: How intensive is it and to what extent does it differ across countries?

• Published date: 01 June 2020
• Date: 01 June 2020
• DOI: http://doi.org/10.1111/ilr.12105

Copyright © The authors 2020

Journal compilation © International Labour Organization 2020

* Institute of Sciences, National University of General Sarmiento, email: labeccari@gmail.com.

** Interdisciplinary Institute of Political Economy of Buenos Aires (UBA–CONICET), email: roxanad

maurizio@gmail.com.

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

publication does not constitute an endorsement by the ILO.

International Labour Review, Vol. 159 (2020), No. 2

Labour market turnover

in Latin America: How intensive

is it and to what extent does

it differ across countries?

Luis BECCARIA* and Roxana MAURIZIO**

Abstract. The study of labour market turnover is particularly signicant in Latin

America, which is characterized by marked economic cycles and limited social pro-

tection coverage. This article estimates the intensity of transitions from employ-

ment in six countries in Latin America in the new millennium and decomposes

the dierences observed, furthermore evaluating the employment destinations of

workers making such transitions. The countries under analysis show very dierent

turnover rates, which are mostly explained by a dissimilar incidence of informal

and temporary employment. In all cases, a large share of job exits imply transi-

tions to precarious jobs or unemployment.

Keywords: labour turnover, informal employment, Latin America.

1. Introduction

The analysis of labour market dynamics considers transitions between jobs,

movements in and out of unemployment and intermittency in the labour force.

It is relevant for a better understanding of the functioning of the labour mar-

ket, evaluating household welfare dynamics and designing public policies. Most

studies of Latin American labour markets are based on analyses of static infor-

mation. Although this is appropriate when investigating factors such as employ-

ment structure or income generation, it only gives a partial overview of changes

in the labour market.

The dynamic approach draws on information on the transitions of people

from, to and within the labour market. One of its signicant aspects is the

study of the eects of these movements on individuals and their households.

For example, the increase in labour market transitions that comes with rising

unemployment levels may have a negative impact on welfare as it broadens

International Labour Review

162

income uctuation and thus increases uncertainty within households. Such

situations are particularly dicult for lower-income families as they tend to be

more vulnerable to social risks. Likewise, employment instability, especially if

it implies transitions to unemployment, can have a negative impact on other

household welfare indicators (aside from income), such as school attendance

and access to welfare programmes, among others. In contrast, some transitions

from unemployment to employment may reduce income uctuation.

Labour market turnover may negatively aect workers’ career paths in vari-

ous ways. First, frequent transitions between jobs can aect individuals’ social

integration and tend to be associated with limited social security coverage. Sec-

ond, high intermittency may prevent individuals from accumulating specic

skills that allow them to increase their level of total human capital. Lastly, the

involuntary loss of a job may make workers less likely to nd future employ-

ment, or employment at the same wage. In such cases, labour market transitions

interrupt the accumulation of skills, with a negative eect on both the worker

and society as a whole.

However, turnover is not always a negative factor or a sign of diculties

in the labour market. Indeed, transitions may be voluntary or reect a move

towards better wages or working conditions. For example, among young

people, they are usually associated with the early stages of their careers, when

the largest movements within the labour market take place as workers attempt

to nd jobs that match their skills or qualications. Labour force intermittency

may be explained by workers’ involvement in other extra-economic activities,

such as education. Furthermore, it may imply virtuous knowledge-sharing

patterns, which have a positive impact on aggregate productivity. Consequently,

high labour market turnover does not necessarily reect appropriate or in-

appropriate labour market behaviour; this can only be determined through

empirical analysis.

This article seeks, rst, to analyse the intensity of exits from employment in

Argentina, Brazil, Ecuador, Mexico, Paraguay and Peru; second, to identify the

most frequent types of transitions and to break down the gaps between the in-

tensities of labour market turnover in these countries into exits explained by

divergences in employment structure and those associated with turnover rate

dierentials between groups of workers; and, third, to identify the destinations

of workers who leave a given job.

This article is relevant because it provides evidence on the rate of labour

market turnover – in other words, transitions from employment – and on the

destinations of such transitions, two variables for which no estimates are avail-

able for Latin America. This is particularly important given the pronounced eco-

nomic cycles and the low levels of social protection that have long characterized

the labour markets in the region.

The selection of countries is based, on the one hand, on the availability of

panel data and, on the other hand, on their relevance in carrying out an ex-

haustive evaluation of the extent of labour market turnover in the region, since

they exhibit widely diering labour market structures and dynamics. They also

represent the largest economies in the region, accounting for approximately

70 per cent of its total population.

Labour market turnover in Latin America 163

The rest of this article is organized as follows. The second section contains a

review of the literature on labour market turnover in Latin America. The third

describes our data sources, while the fourth sets out our methodology. The fth

section examines the composition of employment in the countries under study

and the sixth presents a descriptive overview of job exits. The seventh section

analyses the results of decomposition exercises, the eighth assesses the des-

tinations of transitions and the last section summarizes our conclusions.

2. Literature review

At least ve major stylized facts regarding labour market dynamics emerge from

the international literature: (1) a high proportion of employment relationships

last for a long period of time; (2) most new jobs end very quickly; (3) as a con-

sequence of (1) and (2), a negative correlation exists between the probability

of leaving a job and its duration (Mincer and Jovanovic, 1981; Farber, 1999a);

(4) there are strong discrepancies in labour market turnover according to indi-

vidual or job characteristics; and (5) labour instability has changed signicantly

over time in most countries.

The rst and second of these stylized facts suggest that labour markets are

not “spot markets”, where the employment contracts between workers and com-

panies are adjusted on a daily basis. However, neither are they static markets,

where workers start and end their careers in a single company.1

Unlike in the developed world, few studies have been carried out on labour

market turnover in Latin America. Those that are available have mainly exam-

ined the impact of labour market reforms (especially during the 1990s) on job

exit rates. In this vein, Kugler (2000) found that average tenure of wage employ

-

ees in the formal sector during the 1990s fell more markedly than among the

informal workers who were not directly aected by the labour reforms in Co-

lombia. On this basis, the author concludes that these reforms had a signicant

impact on job instability.

Paes de Barros, Leite Corseuil and Bahia (1999) analysed the eects of the

higher redundancy costs implemented in Brazil in 1998 , nding evidence con-

sistent with the hypothesis that these increases tended to reduce dismissal rates

and thus extend the average length of employment.

Saavedra and Torero (2000) concluded that the labour “exibilization” pro-

gramme that had been implemented in Peru since 1992 had tended to increase

labour market turnover. A more recent study on this country (Herrera and Rosas

Shady, 2003) shows that it is characterized by high job instability, where most

transitions occur between employment and inactivity.

Galiani and Hopenhayn (2000) estimated the probability of workers exit-

ing both employment and unemployment in Argentina during the 1990s. They

found greater instability during the second half of the decade but were unable

to support the hypothesis that the labour reforms implemented in the second

half of the 1990s had had an impact on stability. Beccaria and Maurizio (200 4)

1 See, for example, Farber (199 9a) for the United States.