Versión de HTML Básico
Assessing the Effect of Conditional Cash Transfers in Children Chronic Stunting: The Human Development Bonus in Ecuador
89
13
Analiti a, Revista de análisis estadístico, Vol. 13 (1), 2017
2.3 CCTs and children stunting: evidence
In Latin America and the Caribbean CCTs became popular since mid-nineties, being the
first to implement them Brazil in 1995 with
Bolsa Escola
and
Programa de Erradicacao do
Trabalho Infantil (PETI)
. Subsequently, researches focused on effect estimation came to be
of academic and political concern (Fiszbein et al., 2009). Evaluations on children nutritional
status have shown to be non-conclusive, and sensitive to sample size and estimation method
(Paxson and Schady, 2007).
For example, Attanasio et al. (2005) showed that the Colombian programme
Familias en
Acci´on
, introduced in 2001, improved nutritional status only for children under 24 months
old. Through a Difference-in-Differences (DID) approach from a randomly selected sample
of municipalities from treatment and control groups, the authors estimated a statistically
significant increase in height-for-age z-score or HAZ of 0,161 standard deviations and a
decrease in the probability of stunting of 0,069.
Behrman and Hoddinott (2005) based on a randomised trial found that
PROGRESA
in
Mexico had significant positive effects on nutrition only after implementing child fixed-effects
to control for unobserved heterogeneity. They reported an increase of approximately 1cm in
mean growth per year of the treatment compared to the controls, for kids between 12 and
36 months.
In the Caribbean, a recognised large-scale programme is the Nicaraguan
Red de Pro-
tecci´on Social
, which is focalised in rural areas and started in 2000. Maluccio and Flores
(2004), through a DID approach, applied on an experimental dataset, estimated that the
effect of
Red
was a decrease in the proportion of stunted children under 5, of about 5,3%
and an increase in HAZ of 0,17 standard deviations.
Contradictorily, Morris et al. (2004) in their study of
Bolsa Alimenta¸c˜ao
in Brazil, which
was launched in 2001 as part of a larger programme, estimated that the gain in anthropo-
metric measures is slightly larger for the non-exposed children. The authors reconstructed
a dataset from original administrative registries and individually matched beneficiaries to
excluded households. Their results, which they pair to an intention-to-treat estimate in a
randomised trial, evidence a statistically significant reduction of the mean HAZ of 0,13 sd.
for children under 7 years. Likewise, other countries have evaluated their strategies, though
compared to the amount of researches focused on non-health or nutritional related outcomes,
the number is diminished. Particularly, due to the difficulties to collect these type of infor-
mation and given the intricate biological aspects that affect the analysis (Ranganathan and
Lagarde, 2012).
7