Intra-household inequality in Brazil: using a collective model to evaluate individual poverty

• Ano de defesa: 2016
• Autor(a) principal: Iglesias Pinedo, Wilman Javier
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Viçosa
• Palavras-chave em Português: Renda - Distribuição - Brasil; Família - Aspectos econômicos - Brasil; Pobreza - Brasil; igualdade - Brasil; Crescimento e Desenvolvimento Econômico
• Link de acesso: http://www.locus.ufv.br/handle/123456789/7678

The distribution of resources within households is crucial to the understanding of its members’ material well-being and for the design of redistributive policies. Although the apparent importance of the intra-household dimension of inequality, very little research has focused on how much of the family resources are dedicated to each member, and thereby attempting to assess individual poverty. In fact, the assessment of poverty and inequality often assumes an equal distribution of resources among household members. Moreover, poverty measures not only neglect the distribution of resources within families, but also the gains from joint consumption. However, the share of household resources devoted to each family member is hard to identify, because consumption is measured at the household level and goods can indeed be shared. This research attempted to analyze the extent of inequality within households and its contribution to levels of poverty in the Brazilian context. In particular, we estimated the process of resources allocation and economies of scale in households from Brazil using a collective model of household consumption. More specifically, we attempted to analyze the resource shares of children and adults in relation with the scale economies of joint consumption and the parental bargaining in order to calculate a direct measure of individual poverty for Brazil. The identification of the household member’s resource share requires the observation of adult-specific goods and a joint estimation on couples and singles. This identification strategy differs from the traditional Rothbarth method, in that it is compatible with economies of scale as well as with parents’ bargaining. The database used was Pesquisa de Orçamentos Familiares (POF 2008-2009). The main results provide evidence of inequality and economies of scale within Brazilian households, which leads us to the rejection of the unitary model and the traditional Rothbarth approach for Brazil, respectively. Our findings also showed that men’s share of total expenditures is slightly larger than women's shares for almost all the family structures considered here. The magnitude of children’s shares, interpreted as the cost of children for the parents, is in turn comparatively smaller. We also showed how resources devoted to each household member vary by family size and structure, and we find that, particularly, standard poverty measures tend to overstate the incidence of child poverty. Furthermore, we found that the share of resources devoted to children rises with the number of children, but the average share per child tends to decrease. On the other hand, we found that adult's scale economies are large and affect poverty measures. Specifically, adult poverty is smaller because parents are highly compensated by the scale economies due to joint consumption. Despite that the children command a reasonably large share of household resources, such share is not enough to avoid having higher rates of poverty than their parents. In addition, we found that mothers seem to contribute more resources than fathers to children, and we do not find evidence of gender but age differences in children’s resource shares. Also, results provide evidence indicating that women’s bargaining power within the household improves with their age, level of education and participation in the labor market. Overall, our main finding is that there is substantial intra-household inequality. One important consequence of this is that standard per-capita poverty measures, which by construction ignore intra-household inequality, present a misleading picture of poverty, particularly for children. Finally, our estimates are important for redistributive policy interventions, because they constitute more accurate measures of the relative material welfare of Brazilians in households of varying composition. Furthermore, the fact that it is plausible to measure of each member’s resource shares within households is a very useful step in measuring individual poverty, and thus informing in a more precise way to policy makers which are focused on poverty alleviation.