East Coast Ballroom
The Impact of Bolsa Família Participation on Unhealthy Consumption (306671)
*Fernanda Araujo Maciel, Bentley UniversityKeywords: Bolsa Familia, Brazil, Machine Learning, Propensity Score Matching, Unhealthy Consumption
The objective of this paper is to assess the impact of Brazil's Bolsa Família conditional cash-transfer program on unhealthy consumption, which is measured by expenses with ultra-processed food, alcohol, and smoking products. Applying Machine Learning methods to improve Propensity Score Matching, I analyze the intensive margin, extensive margin, and the overall effect of participating in the Bolsa Família program on the household purchase of unhealthy products. Results show that program participants increase the expenses on food, food away from home, and cookies. Among households that purchase soda or smoking products, the amount spent on these products reduces.