The influence of Bolsa Familia conditional cash transfer program on child labour in Brazil

South America

Brazil

Social Protection

Conditional Cash Transfers (CCT)

Paloma Santana Morei ra Pais, F de Figueiredo Silva, EC Teixeira

2017

Pais, Paloma Santana Moreira, Felipe de Figueiredo Silva, and Evandro Camargos Teixeira. “The Influence of Bolsa Familia Conditional Cash Transfer Program on Child Labor in Brazil.” International journal of social economics 44.2 (2017): 206–221.
View publication

Secondary Data: 2006 National Household Sample Survey. Available at: https://ibge.gov.br/en/statistics/social/income-expenditure-and-consumption

2013

Bolsa Familia (“Family Allowance”)

2004-2006 (beginning of the program up to date on survey data)

Break the poverty trap and general conditions that allow children to enter the job market early.

Children under 17 years of age in low-income families.

Bolsa Familia is a money transfer program subject to two main preconditions. First, it requires compulsory school attendance for children between 6 and 17 years of age. Second, it requires social assistance to be provided to children and teenagers who are up to 15 years old and who have been removed from child labor by the Child Labor Eradication Program. Children who participate in any labor activities should participate in the Service of Living and Strengthening of Links (SCFV in Portuguese) and achieve a minimal monthly attendance of 85 percent.

• School Attendance
• Child labour (hours of work)

Propensity Score Matching

Study design: a non-experimental method (matching) process was used to find a control group with characteristics as close as possible to the treatment group to evaluate the impact of the Bolsa Familia CCT program on child labor. The variables qualifying participation in the Bolsa Familia program and the demographic characteristics used in the logit model were those related to the characteristics of the children and their families. These variables characterize specific social and economic conditions that could contribute to the increased or decreased selection probability for the program. The authors also used the parents’ characteristics and family characteristics. For instance, the father’s schooling and mother’s schooling (in years) was used. The net income per capita from the government transfers aimed to capture the CCT program’s effect.
Sample: The data set used was from 2006 since this year supplemental questions about participants’ access to money transfers and the social programs developed by the government and other institutions were added.
Survey Instrument: The sample was limited to families that had individuals aged between 0 and 17 years and who had a net income per capita (excluding income from government transfers) of up to R$300.00 (US$139.53); the sample totaled 21,886 individuals. This delimitation corresponds to the families who are suitable to receive the Bolsa Familia monetary transfer and thus are families that could be characterized as having a similar level of economic vulnerability.

The program increased the number of hours of child labor in Brazil. However, this outcome might be explained by the fact that those families who received Bolsa Familia were also those with higher socioeconomic vulnerability, which means that they need to guarantee survival with the income generated via child labor.

Child Labour

Households

National

Bolsa Familia increased the number of hours of child labour in Brazil. However, this outcome might be explained by the fact that those families who received Bolsa Familia were also those with higher socioeconomic vulnerability. However, the mother’s schooling, the family’s residence in a metropolitan region, and a higher family net income per capita (excluding income from the government transfer) decreased the probability of child labour after the program was implemented.

-

This study has at least two limitations. First, the data on child labor is only available for 2006, so it is not possible to evaluate the impacts of the program on child labor in other years. Second, since the data is part of a secondary source, the information about child labor could be too low since respondents self-declare and the law forbids this type of work.