Numerous evaluations of conditional cash transfer (CCT) programs show positive short-term impacts, but there is only limited evidence on whether these benefits translate into sustained longer-term gains. This paper uses the municipal-level randomized assignment of a CCT program implemented for five years in Honduras to estimate long-term effects 13 years after the program began. We estimate intent-to-treat effects using individual-level data from the population census, which allows assignment of individuals to their municipality of birth, thereby circumventing migration selection concerns. For the non-indigenous, we find positive and robust impacts on educational outcomes for cohorts of a very wide age range. These include increases of more than 50 percent for secondary school completion rates and the probability of reaching university studies for those exposed at school-going ages. They also include substantive gains for grades attained and current enrollment for others exposed during early childhood, raising the possibility of further gains going forward. Educational gains are, however, more limited for the indigenous. Finally, exposure to the CCT increased the probability of international migration for young men, from 3 to 7 percentage points, also stronger for the non-indigenous. Both early childhood exposure to the nutrition and health components of the CCT as well as exposure during school-going ages to the educational components led to sustained increases in human capital.
Experimental long-term effects of early-childhood and school-age exposure to a conditional cash transfer program
Résumé
en
Numerous evaluations of conditional cash transfer (CCT) programs show positive short-term impacts, but there is only limited evidence on whether these benefits translate into sustained longer-term gains. This paper uses the municipal-level randomized assignment of a CCT program implemented for five years in Honduras to estimate long-term effects 13 years after the program began. We estimate intent-to-treat effects using individual-level data from the population census, which allows assignment of individuals to their municipality of birth, thereby circumventing migration selection concerns. For the non-indigenous, we find positive and robust impacts on educational outcomes for cohorts of a very wide age range. These include increases of more than 50 percent for secondary school completion rates and the probability of reaching university studies for those exposed at school-going ages. They also include substantive gains for grades attained and current enrollment for others exposed during early childhood, raising the possibility of further gains going forward. Educational gains are, however, more limited for the indigenous. Finally, exposure to the CCT increased the probability of international migration for young men, from 3 to 7 percentage points, also stronger for the non-indigenous. Both early childhood exposure to the nutrition and health components of the CCT as well as exposure during school-going ages to the educational components led to sustained increases in human capital.
Auteur(s)
Teresa Molina Millan1
, Karen Macours2, 3
, John Maluccio4
, Luis Tejerina5
1
NOVA SBE -
NOVA - School of Business and Economics
( 111433 )
- Campus de Carcavelos, Rua da Holanda, 2775-405 Carcavelos.
- Portugal
Universidade Nova de Lisboa = NOVA University Lisbon ( 327246 )
2
PSE -
Paris School of Economics
( 301309 )
- 48 boulevard Jourdan 75014 Paris
- France
Université Paris 1 Panthéon-Sorbonne ( 7550 )
;
École normale supérieure - Paris ( 59704 )
;
Université Paris Sciences et Lettres ( 564132 )
;
École des hautes études en sciences sociales ( 99539 )
;
École des Ponts ParisTech ( 301545 )
;
Centre National de la Recherche Scientifique ( 441569 )
;
Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement ( 577435 )
3
PJSE -
Paris Jourdan Sciences Economiques
( 578027 )
- 48 boulevard Jourdan 75014 Paris
- France
Université Paris 1 Panthéon-Sorbonne UMR8545 ( 7550 )
;
École normale supérieure - Paris ( 59704 )
;
Université Paris Sciences et Lettres ( 564132 )
;
École des hautes études en sciences sociales ( 99539 )
;
École des Ponts ParisTech ( 301545 )
;
Centre National de la Recherche Scientifique ( 441569 )
;
Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement UMR1393 ( 577435 )
4
Middlebury College
( 519035 )
- États-Unis
5
Inter-American Development Bank
( 151013 )
- Washington
- États-Unis
Inter-American Development Bank ( 337970 )
Licence
Paternité
Langue du document
Anglais
Nom de la revue
Journal of Development Economics
(ISSN : 0304-3878)
Publié par Elsevier
Revue non référencée dans Sherpa-Romeo
Vulgarisation
Non
Comité de lecture
Oui
Audience
Internationale
Date de publication
2020-03
Volume
143
Public visé
Scientifique
Sous-type de document pour les Articles
Research article
Domaine(s)
Sciences de l'Homme et Société/Economies et finances
I - Health, Education, and Welfare/I.I2 - Education and Research Institutions/I.I2.I25 - Education and Economic Development
I - Health, Education, and Welfare/I.I2 - Education and Research Institutions/I.I2.I28 - Government Policy
I - Health, Education, and Welfare/I.I3 - Welfare, Well-Being, and Poverty/I.I3.I38 - Government Policy • Provision and Effects of Welfare Programs
O - Economic Development, Innovation, Technological Change, and Growth/O.O1 - Economic Development/O.O1.O15 - Human Resources • Human Development • Income Distribution • Migration
Financement
IDB Economic and Sector Work program "CCT Operational Cycles and Long-Term Impacts"RG-K1422French National Research Agency (ANR)ANR-I7-EURE-0001
Teresa Molina Millan, Karen Macours, John Maluccio, Luis Tejerina. Experimental long-term effects of early-childhood and school-age exposure to a conditional cash transfer program. Journal of Development Economics, 2020, 143, ⟨10.1016/j.jdeveco.2019.102385⟩. ⟨halshs-02297704⟩