by Alessandra C.
Over the years, conditional cash transfers (CCT) programs have had a positive effect in many countries in Latin America. In the region they are widely viewed as one of the most promising innovations to date for pursuing an integrated, synergistic strategy for improving the nutrition, health, and education of children. CCT programs in Brazil, Honduras, Mexico and Nicaragua have been some the most successful cases in the world. Both The World Bank and the Inter-American Development Bank have encouraged their adoption in many low-and middle-income countries (Adato, 2010). In El Salvador, one of the major CCT program “Comunidades Solidarias Rurales” (CSR) focuses on improving education and health of children in rural areas. This study aims to evaluate the effect of the CSR cash conditional transfer program on children's education by measuring participation on the CRS program on children's school attendance in rural areas in the country in 2014. The goal of this study is to provide a critical assessment of the effectiveness of this program, and pose some suggestions to policymakers and CSR program coordinators to study its strengths and limitations. The results of this study are important for policymakers in El Salvador because it will help them understand the program's impact in the communities they serve, and whether or not the program is meeting its goals. In the case that this study shows a negative effect or no effect, it would lead policymakers and directors to re-evaluate the significance of investing in CSR and/or make strategic decisions on how to improve it.
There have been two major CCTs in El Salvador, one "Comunidades Solidarias Rurales" (CSR) particularly focuses on families in rural areas of the country with the goal of improving children education and health. The CSR program is a cash transfer program for families in El Salvador living under extreme poverty that started in 2005. The program’s objective is the betterment of living conditions of Salvadoran families living under extreme poverty, focusing primarily on rural areas by expanding their opportunities and providing the necessary resources through improved network of basic services, program developments, and micro credit. CSR includes two types of conditional transfers: an education and a health transfer. Households are eligible for the education transfer if they have a member between six and 15 years of age who has not yet completed primary school, the education transfer is conditional on school enrollment and attendance. The transfer amount is US $15 per month per household for the education or health transfer, and $20 per month if household are eligible for both types. Payments do not vary by the number of eligible children in the household (de Brauw, 2012). When it first started, the program received political championship from the top levels of government; set up inter-institutional mechanism for administrative coordination, and drew on technical and financial support from varied international resources. El Salvador's specific feature, especially its small size combined with an effective national institution with a widespread network and a mandate related to poverty reduction might have favored the implementation strategy of CCT programs (Britto, 2007).
CCTs in El Salvador & The History of CSR
There have been two major CCTs in El Salvador, one "Comunidades Solidarias Rurales" (CSR) particularly focuses on families in rural areas of the country with the goal of improving children education and health. The CSR program is a cash transfer program for families in El Salvador living under extreme poverty that started in 2005. The program’s objective is the betterment of living conditions of Salvadoran families living under extreme poverty, focusing primarily on rural areas by expanding their opportunities and providing the necessary resources through improved network of basic services, program developments, and micro credit. CSR includes two types of conditional transfers: an education and a health transfer. Households are eligible for the education transfer if they have a member between six and 15 years of age who has not yet completed primary school, the education transfer is conditional on school enrollment and attendance. The transfer amount is US $15 per month per household for the education or health transfer, and $20 per month if household are eligible for both types. Payments do not vary by the number of eligible children in the household (de Brauw, 2012). When it first started, the program received political championship from the top levels of government; set up inter-institutional mechanism for administrative coordination, and drew on technical and financial support from varied international resources. El Salvador's specific feature, especially its small size combined with an effective national institution with a widespread network and a mandate related to poverty reduction might have favored the implementation strategy of CCT programs (Britto, 2007).
The Study Design
The Study Design
I plan to answer this question by obtaining data
from the Latin America Public Opinion Project (LAPOP)- Americas Barometer on El
Salvador, for the year of 2014. LAPOP is the premier academic institution carrying out
surveys of public opinion in the Americas, with over thirty years of
experience. I consider this data to be not only the most relevant, but also the
most up to date and accurate. The study is based on random sampling of 1,512
individuals in 2014. The sample consists of five strata representing the five
main geographical regions of El Salvador. The same is further stratified by
size of the municipalities and by urban and rural areas. For the purpose of this study, I only use the data on individuals who live on rural parts of the country, because only individuals residing in rural areas are eligible for the CSR program benefits, which would make the final simple size of 474.
I am using children's education as the dependent variable, measured by whether or not the respondent had any children attending school in the past year. I recoded the variable so it does not take into account if the respondent’s children did not go to school for the following reasons: he/she did not have any children, did not have children old enough to go to school, there was no more space in the school, or if the school was too far away. I am measuring the independent variable by whether or not the respondent is a participant of the Comunidades Solidarias Rurales program. Additionally, I am controlling for confounding variables that could influence the effect of both children’s school attendance and participation in the program. The variables I am accounting for are: whether or not respondents are participants of other CCTs, recipients of remittances, or recipient of government assistance. I am also controlling for mother's education level and income. For a further variable description see the following table.
Dependent, Independent and Confounding Variables
I am using children's education as the dependent variable, measured by whether or not the respondent had any children attending school in the past year. I recoded the variable so it does not take into account if the respondent’s children did not go to school for the following reasons: he/she did not have any children, did not have children old enough to go to school, there was no more space in the school, or if the school was too far away. I am measuring the independent variable by whether or not the respondent is a participant of the Comunidades Solidarias Rurales program. Additionally, I am controlling for confounding variables that could influence the effect of both children’s school attendance and participation in the program. The variables I am accounting for are: whether or not respondents are participants of other CCTs, recipients of remittances, or recipient of government assistance. I am also controlling for mother's education level and income. For a further variable description see the following table.
Findings
The results of cross tabulation and multivariate regression (see Tables 2 and 3) suggests that children of participants of the Comunidades Solidarias Rurales program, are 20.5% less likely to attend school. The data portrays with a 99.9% certainty that on average, for every child of a participant of the CSR program there is a .205 point decrease in school attendance; controlling for all other variables in the model. It is safe to believe that this is not due to chance considering that the p-value<0.01 (see Table 3).
Data Limitations
The data used in this analysis has potential limitations. ]The data is limited to children attending school in the year prior to 2014, thus it is unknown how whether some children had attended school years prior to that. Additionally, the data does not reflect what percentage of children in rural area do not attend school in general. Perhaps it could be that overall, children in these parts of the country have a lower school attendance, thus, inferring a problem of reverse causality. Which would mean that despite its negative relationship shown in the Table 3, the CSR program is in fact is improving children's school attendance. Thus, the true effect of the CSR program could be underestimated.
Conclusion: There is room for improvement
Children's school attendance in rural areas of El Salvador are still low. Although the Comunidades Solidarias Rurales program focuses on improving children's education by conditioning cash transfer to school attendance, we still see an overall decline of school attendance in rural areas. As mentioned above, this could be due to the fact that in general, parents in these parts of the country are less likely to send their children to school. However, these results reflect two main principles: 1) that policymakers and government officials need to increase their focus and investment into programs that work to improve children's education in rural areas. 2) Directors/coordinators of the CSR program need to re-evaluate its implementation and goals, and if needed be work towards making significant chances to improve the execution of the CCT program. Nevertheless, this findings be used to shed light into the betterment of the education system in the country and the management of the CSR program and other conditional cash transfer programs in the future.
References
De Brauw, A. (2012). Regression Discontinuity Impacts with an Implicit Index: Evaluating El Salvador's Comunidades Solidarias Rurales Transfer Programme. International Policy Centre for Inclusive Growth. Brasilia, DF. Brazil.
Adato, M., Hoddinott, J. (2010). Conditional Cash Transfers in Latin America. The Johns Hopkins University Press. Baltimore, MD, U.S.A.
Comments from Heaven F:
ReplyDeleteI appreciate the background you provided on the cash program, I think your independent variable confounding variables are clear. I also like the reasoning behind your data choices. I'm curious about the header "the import", maybe just the CSR program would suffice? Your control variables are great, and encompass some of the factors in other confounding variables. Once you provide more analysis the reader will have a better picture. I'm sure you are aware of this, but the graphics of the regression table might be a bit blurred, maybe because they are in bold. I think you have a great start!
The research question of this post was clear although I did find myself having to go back to confirm what the DV and IV were. It might help if the variables in the tables had names a reader could follow or easily identify. The background on the program was very helpful but did raise a few questions in terms of the research design. For example, were people not participating in the program eligible but chose not to participate or were they persons who were not eligible and thus had higher incomes? The latter may explain the results of the regression as persons with higher incomes can more easily afford to send children to school. This would require discontinuity method we discussed in class where only persons who fall close to the maximum income line on either side are considered. It is also not clear that persons who list themselves as participating in the program are participating under the education transfer because they could be participating under the health transfer. If all this is controlled for, the result observed may still be valid. Consider that, if a family is very poor, the money they save by not sending kids to school and the additional income from instead sending that child to work may be larger than the subsidy. Thus, the subsidy is not enough to keep kids in school.
ReplyDelete