Governments increasingly use conditional cash transfer (CCT) programs as part of their poverty alleviation strategy. These programs provide poor families with cash subsidies contingent on adherence to certain activities, with the hope of limiting the negative effects of poverty on children’s development. While much is known about the benefits of CCTs for children’s nutrition, physical development, and school enrollment, very little is known about how CCTs can affect children’s cognitive and behavioral outcomes. Theory suggests that cash transfer programs may positively impact child behavior through improved nutrition or by mitigating severe economic stress on the family, which would enable parents to provide more consistent structure and monitoring for their children. These effects can be particularly important for children four to five years old; their ability to self-regulate and develop successful relationships with adults and peers at this age strongly predicts later educational achievement, emotional maturity, and delinquency. Such theoretical impacts of CCT programs have, until now, been untested. This study is the first to examine the empirical impact of CCTs on children’s behavior problems.
The Mexican government began the well-known CCT program Oportunidades (originally called Progresa), in 1997. Oportunidades offers a bimonthly transfer worth 25 percent of household income (equal to US $15 for the poorest households) conditional on behavior changes such as use of prenatal care services, well-infant care and immunization, nutrition monitoring and supplementation, preventive checkups, and participation in health educational programs. The program targeted impoverished rural areas in Mexico in attempt to relieve extreme poverty.
The primary outcome measure for this study was maternal reporting of behavior problems in terms of anxiety/depressive and aggressive/oppositional symptoms. All data were gathered via home interviews by teams of trained nurses from Mexico’s Instituto Nacional de Salud Publica. The treatment sample was comprised of 506 poor rural communities that had received Oportunidades benefits every two months for three and a half to five years.
A comparison group of 152 non-Oportunidades communities was created using propensity score matching (PSM), a quasi-experimental method used retrospectively to imitate randomization. Additionally, the control group was further restricted to families that were subsequently offered and enrolled in Oportunidades by 2005, which was done to minimize the likelihood that factors other than treatment status would differentiate the two groups.
Results and Policy Implications
Bimonthly transfers of approximately $15 were associated with a 10 percent decrease in aggressive and oppositional symptoms in children, after controlling for a range of child characteristics and socioeconomic factors. There was not, however, any statistically significant program effect found for symptoms of anxiety and depression. None of the program effects found through the study varied by the child’s gender or ethnicity.
This study demonstrates that CCT programs can positively affect outcome measures other than those on which the cash subsidy is specifically conditioned. While more research is needed to better understand the relationship between CCTs and child behavior, these findings suggest that interventions aimed toward increasing basic human capital may exert longer-term positive ripple effects on childhood development outcomes. Moreover, extensive research on behavioral problems in high-income countries suggests that even small differences early in development can magnify over time to influence a child’s long-term developmental trajectory.,
 Greenberg MT, Domitrovich C, Bumbarger B. The prevention of mental disorders in school-aged children: current state of the field. Prev Treat. 2001. Available at: http://psycnet.apa.org/, Accessed January 5, 2006
 Flannery DJ, Vazsonyi AT, Liau AK, et al. Initial behavior outcomes for the PeaceBuilders universal school-based vio- lence prevention program. Dev Psychol. 2003;39(2):292–308