Demanda domiciliar por alimentos ultraprocessados no Brasil: uma análise dos dados da POF 2017-2018

• Ano de defesa: 2021
• Autor(a) principal: Moura, Marielly Marie da Silva
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Economia; Programa de Pós-Graduação em Economia; UFPB
• Palavras-chave em Português: Saúde pública - Economia; Alimentos ultraprocessados; Pesquisa de Orçamentos Familiares - POF; Modelo Quaids; CNPQ::CIENCIAS SOCIAIS APLICADAS::ECONOMIA
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/123456789/27464

The increase in sales and, consequently, in the consumption of ultra-processed foods has been gaining prominence on the international agenda of debates on public health and the economy. Specifically due to associations between the consumption of these foods and increases in the relative risk of individuals developing non-communicable chronic diseases, such as obesity, diabetes, cardiovascular diseases, among others (PAHO, 2018). In the economic context, these diseases bring costs to societies and governments, which can be measured in terms of lost productivity at work, premature mortality and expenses with medical and hospital care (COLDITIZ, 1992). Therefore, this study aims to analyze the demand of Brazilian households for ultra-processed foods, based on the use of microdata from the new Family Budget Survey (POF/IBGE 2017-2018). For this purpose, a demand system with fifteen food groups was estimated through the two-step estimation procedure proposed by Shonkwiller and Yen (1999). The approach used was the Quadratic Ideal Demand System (Quaids) with adjustment for possible econometric problems. In the first stage of estimation, equations were estimated using the probit model to verify which factors influence household decisions to purchase or not a particular food. In the second stage, it was possible to analyze the sensitivity of consumer demand to changes in food prices, as well as changes in restricted total expenditure. The results indicate that demand sensitivity varies according to income classes and geographic regions. As for the total expenditure, it was found that the degree of sensitivity in the ultra-processed category increases with the level of income. Specifically, for foods in the sweet and savory cookies group, the elasticity-expenditure value of demand in the poorest households is 0.33 and for the richest households this value jumps to 0.61, that is, the poorest households (1st income quartile) tend to be more sensitive to changes in total expenditure. Regarding the regions, it was observed that cheeses, cold cuts and sausages are more sensitive to variations in the total expenditure in the north and northeast regions; cakes and sweets in general in the southeast and industrialized breads to the north and south. Furthermore, in relation to prices, demand from Brazilian households was elastic for almost all product groups in the ultra-processed food category. Thus, for policy purposes aimed at discouraging the consumption of these foods, the results suggest that increases in the prices of foods in this category would lead to a proportionally greater reduction in the quantity demanded than the increase in prices.