Migration, Transfers and Child Labor

• Author: Ralitza Dimova,Ira N. Gang,Gil S. Epstein
• DOI: http://doi.org/10.1111/rode.12156
• Date: 01 August 2015
• Published date: 01 August 2015

Migration, Transfers and Child Labor

Ralitza Dimova, Gil S. Epstein, and Ira N. Gang*

Abstract

We examine agricultural child labor in the context of emigration, transfers and the ability to hire outside

labor. We start by developing a theoretical background and show how hiring labor from outside the house-

hold and transfers to the household might induce a reduction in children’s working hours. Analysis using

Living Standards Measurement Survey (LSMS) data on the Kagera region in Tanzania lend support to the

hypothesis that both emigration and remittances reduce child labor.

1. Background

According to the International Labor Organization (ILO–IPEC, 2013) about 168

million children between the ages of 5 and 14 are working in developing countries.

Asia and the Pacific have the largest numbers (almost 78 million or 9.3% of the child

population), but Sub-Saharan Africa continues to be the region with the highest inci-

dence of child labor (59 million, over 21%). There are 13 million (8.8%) children in

child labor in Latin America and in the Caribbean and in the Middle East and North

Africa there are 9.2 million (8.4%). The negative implications of even the less hazard-

ous occupations on human capital accumulation, child and adult health and future

outcomes such as labor market performance and intergenerational poverty traps is

well documented (Basu, 1999, 2000; Basu and Chau, 2008; Edmonds, 2008).

The literature explaining child labor incidence in developing countries centers

around (i) the need for child employment to meet the family’s very basic subsistence

requirements (Basu and Van, 1998; Basu, 1999), (ii) a response to the absence of

credit markets (Baland and Robinson, 2000; Ranjan, 1999, 2001), and (iii) low returns

to education (Bacolod and Ranjan, 2008). Child labor generally falls with increases in

gross domestic product (GDP) per capita, life expectancy, international trade, educa-

tional spending and GDP growth (Saad-Lessler, 2010). Policy proposals include the

universal banning of child labor and banning the use, export and imports of products

made with child labor, improving credit markets, imposing minimum wage restrictions

and providing income support to households. The literature examines other policy

instruments including food for education and investment in education (Jafarey and

Lahiri, 2005) and, of course, trade sanctions (Jafarey and Lahiri, 2002). No single

approach has managed to provide a tangible solution to the problem.

Recent research has extended the stylized gamut of explanations and policy recom-

mendations in innovative directions. Taking as a starting point the view that child

labor is a result of excess labor supply and subsistence constraints (Basu and Van,

1998), Epstein and Kahana (2008) show that reducing the amount of labor available

* Gang: Economics Department, Rutgers University, 75 Hamilton Street, New Brunswick, NJ 08901-1248,

USA. Tel: +1-848-932-8648; Fax: +1-732-932-7416; E-mail: gang@economics.rutgers.edu.Also affiliated to

IZA, Bonn, CReAM, London and IOS, Regensburg. Dimova: Institute of Development Policy and Man-

agement, University of Manchester, Manchester, UK. Also affiliated to IZA, Bonn, Germany. Epstein: Bar

Ilan University, Ramat Gan, Israel. Also affiliated to IZA, Bonn, Germany and CReAM, London, UK.

Review of Development Economics, 19(3), 735–747, 2015

DOI:10.1111/rode.12156

© 2015 John Wiley & Sons Ltd