The continuous treatment effect of an antipoverty program on children's educational attainment: Colombia's Familias en Accion

• Date: 01 August 2018
• Published date: 01 August 2018
• Author: Juan M. Villa
• DOI: http://doi.org/10.1111/rode.12380

REGULAR ARTICLE

The continuous treatment effect of an antipoverty

program on children’s educational attainment:

Colombia’sFamilias en Accion

Juan M. Villa

Global Development Institute, University

of Manchester, United Kingdom

Correspondence

Juan M. Villa, The University of

Manchester, Arthur Lewis Building,

Oxford Road, Manchester M13 9PL,

United Kingdom.

Email: juan.villa@manchester.ac.uk

Abstract

Educational attainment is an important element in the for-

mation of human capital. Although many developing coun-

tries have made strong efforts to expand the coverage of

education services, children in poor households still strug-

gle to attend school on a regular basis. Human develop-

ment conditional cash transfers (known as CCTs) have

emerged in response to this situation in developing coun-

tries. While the effects of the CCTs are well known and

widely documented, their effects in relation to educational

attainment and school participation are still unclear. This

paper looks empirically into the continuous treatment

effects of participation length in Familias en Accion,a

CCT program in Colombia. The paper focuses on the con-

tinuous treatment effects on school registration and educa-

tional attainment of participants in the program. Although

initial results show a fuzzy relationship between the pro-

gram outcomes and the participation length, the empirical

results confirm the fact that a longer exposure to the anti-

poverty program led to higher school registration rates,

accumulation of years of education, and lower child labor

participation levels.

1

|

INTRODUCTION

School attendance and the attainment of higher levels of education are crucial inputs in human

capital formation. Although many developing countries have made important efforts to expand the

DOI: 10.1111/rode.12380

Rev Dev Econ. 2018;22:1239–1262. wileyonlinelibrary.com/journal/rode ©2018 John Wiley & Sons Ltd

|

1239

coverage of education services by building schools, providing free basic education and facilitating

universal access to schools, many impoverished households still struggle to send and keep their

children at school. As children may work and actively contribute to household income, their par-

ents at times are forced to choose between subsistence food consumption and sending their chil-

dren to school. This creates a reinforcing cycle of low education and poverty (World Ban k, 2000).

Consequently, policy has responded with the rapid introduction of particular antipoverty programs.

These programs have the dual goals of alleviating poverty while also making children attend

school on a regular basis. To achieve both, more than 60 countries around the globe have adopted

human development conditional cash transfer programs (CCTs). These antipoverty programs intend

to increase investment in the human capital of children belonging to poor households by delivering

income in the form of cash (Barrientos, 2013).

CCTs are the most popular antipoverty intervention in Latin America. They have also been

implemented in some African and Asian countries. To date, there are more than 100 million poor

households participating in these programs (Barrientos & Villa, 2015). CCTs provide cash transfers

to adults in poor households conditional on their children’s maintained school attendance and

health check-ups (Attanasio, Oppedisano, & Vera-Hernandez, 2014). These conditions placed on

the transfers, disguised as co-responsibility, are intended to improve the nutritional status of indi-

viduals in early childhood and increase the years of education attained by school aged children.

The main rationale of CCT programs is that the increased human capital formation during one’s

youth will lead to a higher productive capacity in adulthood.

One of the main sources of support for the adoption of CCTs in many countries is a history of

positive impact evaluations. There is plenty of evidence concerning the effectiveness of CCT pro-

grams in improving a variety of dimensions through intended and unintended results. Exa mples

include nutrition and school attendance (Rawlings & Rubio, 2005), test scores (Baez & Camacho,

2011), grade repetition (De Oliveira & Soares, 2013), income, savings, consumption, and labor

supply (Skoufias & Gonzalez-Cossio, 2008), cognitive abilities (Ponce & Bedi, 2010), fertility and

gender issues (Soares & Silva, 2010), agricultural outcomes (Todd, Winters, & Hertz, 2010), intra-

household allocation (Del Carpio & Macours, 2009), economic growth (Villa, 2016) and poverty

and inequality (Bourguignon, Ferreira, & Leite, 2002; Soares, Os

orio, Soares, Medeiros, & Zepeda,

2007). The effectiveness of CCTs in the past and the dissemination of a significant number of pos-

itive evaluation reports have contributed to their adoption in most of the countries in Latin Amer-

ica, Africa, and Asia (Barrientos & Villa, 2013).

Previous studies evaluating antipoverty transfer programs have relied on the static comparison

of treated and untreated groups in a binary setting. The main focus has been placed on the average

effect (Ravallion, 2007). In impact evaluation terms, the effect of the treatment has been consid-

ered as static and not continuous. Continuous treatment effects are defined here as the changes in

observed outcomes at different lengths of participation by beneficiary households in antipoverty

programs. Participation lengths are considered either potential or elapsed. Potential participation

length is defined as the maximum period that beneficiary households can benefit from the pro-

gram. Later it will be shown that potential participation length is determined by the age or school

grade of the youngest child in each household. Elapsed participation is defined as the actual partic-

ipation length measured between the enrollment date and the date at which the household is re-

contacted for a follow-up observation. Here, continuous treatments focus on the elapsed participa-

tion length of the household relative to its potential participation length.

Few efforts to assess the continuous treatment effects of antipoverty transfer programs have

been made. In the field of child nutrition, Behrman, Cheng, and Todd (2004) assessed the opti-

mum time exposure of children to child care centers, considering their mothers’behavior.

1240

|

VILLA