Versión de HTML Básico
Lorena Moreno
92
Analiti a, Revista de análisis estadístico, Vol. 13 (1), 2017
conditions is a challenge for the government, and only recently more accurate actions have
been taken. Nevertheless, beneficiaries have always been informed of the importance of the
co-responsibilities as a requirement for the monthly transfer. Moreover, several national
surveys and public services directly ask recipients about their compliance.
3.2 Motivation
In the same way as with other interventions, the BDH has been evaluated in the light of
its objectives; though, researches have mainly focused on school attendance, child labour,
household consumption and labour indicators as outcomes (Gonzalez-Rozada and Llerena,
2011; Oosterbeek et al., 2008; Ponce, 2011; Rosero and Mart´ınez, 2007; Schady, 2006). While,
fewer studies have investigated children cognitive development and health indicators (Pax-
son and Schady, 2007; Ponce and Bedi, 2010) and even less, only malnutrition (Le´on and
Younger, 2007). The estimates, which correspond to the first stages of the BDH, point
towards a slightly modest to null effect of the treatment on height-for-age.
As mentioned, there are only two relevant specific studies. First, Le´on and Younger
(2007) measured the impact of
Bono Solidario
using the 1999 Living standard Measurement
Survey (LSMS) through and instrumented OLS with various controls. They modelled the
differential effect between regular income and the cash transfer. The instruments were: a)
an interaction of the three programme eligibility criteria (a dummy variable with value 1 for
households with a monthly income less than 150US$, no workers in the formal sector and
at least one mother with a child under 18, and 0 otherwise) b) a continuous variable of the
traveling time to the collection point and c) a dummy with the value of 1 if the interview
took place before April 1999, when the transfer amount was increased and 0 otherwise. They
estimated that there is a significant though modest effect on children nutritional status both
stunting (height-for-age) and underweight (weight-for-age) no different from a regular income
effect.
The second study, by Paxson and Schady (2007), took advantage of the randomized
introduction of the BDH, and computed the Intention-to-treat estimates for kids of ages 3 to
7 by the end of the period (baseline survey was collected October 2003 - September 2004 and
follow-up September 2005 - January 2006). The authors estimated a 4 to 5 percent increase
of a standard deviation (depending on model specifications) in the HAZ of 1.479 children
(corresponding to 1602 nuclear families) in rural areas of six provinces. As mentioned one
of the main reasons for this reduced number of researches is the lack of anthropometric
information related to the transfer recipients, due to the costs it represents.
On top of these limited results, the discouraging evolution of an approximate 2% decrease
in 8 years, from 25,6% in 2006 to 23,9% in 2014 (INEC, 2006, 2014) in national stunting
prevalence, configures a permanent concern of the public, private and non-governmental sec-
tors in Ecuador. Therefore, the measurement of the effect of CCT on children nutritional
status is of great importance not only for theoretical matters but also to narrow the men-
10