How and When: Disentangling Cash and Care Effect of CCTs on Birth Outcomes

INEQUALITY AT BIRTH

Conditional Cash Transfers Reduce the Incidence of Premature Births,

Especially through an Adequate Prenatal Care

Additional resources during pregnancy can improve birth outcomes, but effects are concentrated among white and educated poor mothers, and among those who comply with health services conditions

SOURCE:Anderson Schneider, VEJA

Endowments at birth are key determinants of several short and long run socioeconomic and health outcomes. Importantly, the nine months in utero are a critical period for setting these initial conditions. This suggests that public policies aiming to improve children’s life and fight poverty should take place even before these children are born.

One such policy are conditional cash transfers (CCTs) programs targeted to low socioeconomic status families. Recent research has shown that CCTs have been very successful in improving children’s education and health outcomes, but little is known about how and when CCTs transfers improve in utero conditions of babies. Moreover, most previous research has investigated the impact to CCTs along the extensive margin (program participants vs. non-participants), and there is lacking evidence on whether program expansions within existing beneficiaries can exhibit additional returns.

CCTs can influence birth outcomes both through the cash transfer itself – improving nutritional conditions of the mother and the baby during pregnancy – as well as prevent complications during childbirth – through an adequate management of prenatal care visits. This project analyzes Brazil’s CCT program Bolsa Família, one of the largest CCT program in the world. Bolsa Família is a targeted anti-poverty program that reaches nearly 15 million families and 45 million individuals, 25% of the country’s population.

Our empirical strategy takes advantage of quasi-experimental variation of Bolsa Família eligible transfers – based on rules that depend on family composition and family member dates of birth – to recover the causal effects of the CCT program.

Research Paper available on request

Data Paper available on request

Data Dictionary available on request

KEY FINDINGS

[Data Linkages] The analysis dataset is based on the linkages of several administrative datasets: registry data on families and individuals targeted for welfare government programs in Brazil (Cadastro Único); monthly payroll data on Bolsa Família with all transfers received; information on the conditionality compliance of beneficiaries (eg., prenatal care visits); natality files with birth outcomes; and mortality files for infant mortality.

We construct a longitudinal dataset of children born in low income families, with monthly information on welfare transfers received, their birth outcomes, as well as time-varying family composition and characteristics. The dataset, which we refer to Born in Bolsa Família Longitudinal Data comprises over 6 million births occurring to families in the Unified Registry for Social Programs between 2011 and 2015. Our deterministic linkage algorithm minimizes measurement error and reaches as a positive predictive value of at least 98.7%.

[Causal Analysis] Dates of birth of family members are as good as random in determining the transfer amount and are not subject to choice or manipulation. The feasibility of this design takes advantage of program eligibility rules, as well as data availability at a very granular level, including family member dates of birth and longitudinal information on monthly transfers received, as well as a large sample size, that affords us statistical power to identify very small effects. Our empirical strategy allows us to determine the window of opportunity where CCT interventions exhibit highest impacts on birth outcomes, recognizing heterogeneous impacts according to how early in the pregnancy the CCT intervention starts.

Bolsa Família transfers vary according to family composition and dates of birth for family members. Families with an adolescent who turns 18 after December 31st of a given year receive payments (youth benefit) for one extra year as compared to families with adolescents who turn 18 right before that. Likewise, families for which the youngest member turned 7 before of May 15th of 2012 qualify for an addition 6 month benefit (extreme poverty benefit) as compared to families where the youngest member is slightly older than that. We explore the timing of childbirth within those families to generate exogenous variation in income received during pregnancy.

[Sizable Transfers] Exogenous variation in income received during pregnancy is as much as 25% of total transfers and 64% on monthly income.

[Small Effects of Cash] Large amounts of additional cash transfers are associated to negligible effects on birth outcomes, including birth weight, gestational length and the incidence of preterm births, low birth weight and small for gestational age.

[The Importance of Health Conditions] But small amounts of cash can reduce inequality at birth when coupled with adequate prenatal care. The incidence of preterm birth reduces by 10% (1 percentage point) among mothers meeting health conditions compliance.

[Complementarities] Effects are concentrated in high SES women: white and educated.

[When] Transfers occurring in 1^st trimester impact preterm birth whereas transfers in the 3^rd trimester impact birth weight

WHAT DOES THIS MEAN FOR POLICY?

Results suggest that a promising way to reduce inequalities at birth involves the provision of adequate health services to pregnant mothers living in poverty. Cash transfers improve birth outcomes only when coupled with an adequate prenatal care, and should start early in pregnancy.

Cash transfers to the poor operate beyond overcoming food insecurity and nutrition deprivation for families in welfare, as we detect negligible effects on the incidence of low birth weight. Additional cash improves birth outcomes mainly through the reduction in preterm birth, and is consistent with complementary effects of cash and important health services provided during gestation.

Our findings also speak to the need of additional attention and resources to the black population and the less educated, as CCTs interventions are less effective precisely for those how need it the most.

Need for long-term outlook:

These findings have important implications for the design and expansion of safety net programs that aim to protect vulnerable population of economic downturns and to improve social mobility. It highlights that an important direction for policy expansion are pregnant mothers in poverty. Moreover, the improvement of in utero conditions of babies can have lasting effects over their lives, with higher cost-benefit returns relative to interventions latter in life.

A Path Forward:

We look forward to partnering with organizations and researches engaged in understanding long run effects of the safety net on social mobility of babies born in poverty and to build a networks that can lead to scalable ways to improve opportunities and reduce inequalities for those children. Be in touch: machadoc@gmail.com