Summary of the impact

Marco Manacorda's research on the causes of child labour and its interaction with schooling has influenced policy and led to a more nuanced understanding of the issue by governments and international organisations. Specifically, his work has:

1. provided a method of standardising statistics from different sources, yielding more reliable cross-country comparisons;
2. devised criteria for the evaluation of interventions carried out by the US Department of Labor under its Education Initiative (EI); and
3. enhanced public understanding of the issue of child labour and the efficacy of different policy responses.

Underpinning research

In less developed countries, reducing child labour and increasing school attendance and achievement are essential for long-term growth and the eradication of poverty. However, neither the most effective way of achieving these goals, nor even the exact causal relationship between child labour and schooling is immediately clear.

Manacorda has contributed to the understanding of these issues and to the policy debates surrounding child labour in a series of studies (conducted jointly with colleagues at the World Bank, UCL, and Washington University in St. Louis) funded by a Nuffield Foundation New Career Development Fellowship (2002-05) and a research grant from the ESRC (2002-04).

Manacorda's research on child labour has offered new insights and at times challenged facile assumptions. Key findings, established empirically, include the following:

1. Contrary to a widespread perception, making schools more accessible in poorer countries does not reduce child labour, though it does increase school attendance. These effects were shown using data from rural Tanzania, an area with an agriculture-based economy and a historically high incidence of child labour (see reference #1 in Section 3 below).
2. Within the household, whether or not children work has no discernible effect on the labour supply of either parent. If one child works, however, any siblings are considerably less likely to work and more likely to attend school. This indicates that the gains to the household from child labour are largely redistributed among the children, and hence extreme poverty is a better explanation of the phenomenon than pure exploitation of children by parents. These effects were shown using data from the US census of 1920 (see reference #2 in Section 3 below).
3. While a policy of requiring very poorly performing pupils to repeat school years (and thus banning "social promotion") may create desirable incentives to avoid failure, such a policy is also harmful in that it exerts a causal effect on increased dropout rates and premature participation in the labour market. This is true even when the decision to drop out occurs years after the mandated repetition. These effects were shown using data from junior high schools in Uruguay (see reference #3 in Section 3 below).
4. Pre-school attendance exerts a similarly long-lasting positive causal effect on duration of schooling and scholastic achievement. By age fifteen, having attended pre-school is associated with a 27-percentage point increase in the probability of remaining in education. This suggests that early intervention policies to subsidise or otherwise encourage pre-school participation are an effective way to improve the life chances of children who are at risk of low educational attainment and early entry into the labour market. These effects were shown using data from a period of rapid expansion in pre-school attendance in Uruguay after 1995 (see reference #4 in Section 3 below).

References to the research

Details of the impact

Manacorda's research on child labour has led to a series of collaborative projects with Understanding Children's Work (UCW), a joint programme of the International Labor Organization (ILO), Unicef, and the World Bank. His association with UCW has afforded privileged access to policymakers and statistical offices, and has given him many opportunities to influence practices and debates relating to child labour eradication.

Specific areas of impact from Manacorda's research include the following.

1. Standardisation of statistics.

The mission of the UCW programme includes the collection, archiving, and distribution of accurate statistics on child labour from around the world. These statistics are used heavily by policymakers, development specialists, and the media to measure progress in this area and to make cross-country comparisons. Standardisation of figures arising from different sources is necessary if these measurements and comparisons are to be meaningful.

Manacorda has devoted considerable attention to this problem in his work with UCW. He led a study of the comparability issue (see source #2 in Section 5 below) that came to pessimistic conclusions, finding the child labour statistics (and hence the resulting international rankings) to be worryingly fragile due to variations in the survey instruments used to collect them. He then proceeded to develop a method of standardising these statistics that is scalable and easy to use, providing a valuable service to research and public policy related to child labour incidence.

2. Public sector practices.

Since 2001 the US Department of Labor (USDOL) has distributed approximately $250 million through its Education Initiative (EI) to projects aimed at reducing child labour. In 2009 Manacorda joined the EI's Technical Advisory Group, charged with the task of developing methods to analyse the effectiveness of particular interventions and improving the transparency of the programme as a whole.

Manacorda's recommendations — drawing on methods used in his published research — were made in technical reports (see source #3 in Section 5 below) that outlined the general principles of programme evaluation, highlighted how randomised controlled trials (a powerful empirical strategy commonly used in applied microeconomics) were essential for establishing programme gains in a scientifically credible way, established criteria for assessing existing projects and explained how an evaluation component could be built into the design of new programmes.

A recent analysis of an EI project in Nepal, providing schooling assistance to a group of child carpet weavers, which includes a randomised controlled trial, was shown to confer substantial benefits on both the children and the households in which they lived.

3. Child labour policy debates.

Manacorda is committed to improving understanding of the phenomenon of child labour by governments, international organisations, and the general public. He has taken every opportunity to explain how the tools of applied microeconomics can address this subject and help to guide the formulation of effective policy. For example, he has

• been interviewed by the BBC and Al Jazeera on the occasion of June 12^th, designated the World Day Against Child Labour;
• spoken at practitioner conferences (eg, the Oslo High Level Group on Education for All) and at seminars at international organizations (e.g. the World Bank) on topics such as the interaction between child labour and globalisation, and school accessibility and child labour; and
• participated in the UK consultation on "Vulnerabilities" for the UN Development Programme's Human Development Report 2014.

Manacorda's research in this area has featured in the Lancet World Report on Child Labour as well as in policy documents from organisations including the ILO, UN Development Programme, UNESCO, Unicef, and World Bank (see sources #4-6 in Section 5 below).

Sources to corroborate the impact

The following individual can address the impact of Manacorda's work as a member of the Education Initiative's Technical Advisory Group:

1. Coordinator, Understanding Children's Work Programme: www.ucw-project.org

Technical reports prepared for USDOL and UCW:

2. "Towards consistency in child labour measurement: assessing the comparability of estimates generated by different survey instruments." (With L. Guarcello, I. Kovrova, S. Lyon, and F. Rosati.) June 2010. Available at: http://goo.gl/n2s6Ph
3. (a.) "Criteria and guidelines for the evaluation of the impact of child labour interventions." February 2009. (b.) "Technical criteria for the evaluation of USDOL- funded child labour education initiative project." July 2009.

Other corroborating sources include:

4. "Joining forces against child labour: inter-agency report for The Hague Global Child Labour Conference 2010." ILO/UCW, May 2010. [See discussion of Manacorda's research on grade retention on p. 45 and discussion of Kondylis and Manacorda's research on school attendance on p. 83.] Available at: http://goo.gl/fxvsh2
5. "Social protection in eastern and southern Africa: a framework and strategy for UNICEF." United Nations Children's Fund, 2008. [See discussion of Kondylis and Manacorda's research on school attendance on p. 44.] Available at: http://goo.gl/FTnr4l
6. "Aportes para la elaboración de propuestas educativas." UNESCO, 2009. [See discussion of Manacorda's research on grade retention on p. 215.] Available at: http://goo.gl/ncJhnj