Gésia Coutinho Marcelino

The objective of this research was to measure the magnitude of multidimensional and income poverty in Brazil, using the microdata from the National Household Sample Survey (PNAD), for the period from 2005 to 2015 and for the year 2019. Being the objective of this last period, inclusive, to determine the probability of poverty from the same point of view. In view of the spatial inequalities existing in the Brazilian states, these analyzes have extended to the rural and urban areas of the country. There are three distinct methodological approaches used in this dissertation: the one-dimensional (proportion of poor); the multidimensional (Alkire - Foster method); and the binary response (logit model). The main results obtained for Brazil show that there was a drop in the adjusted proportion of poor (M₀), as well as in the incidence (H) and intensity (A). In 2005 and 2015 the IPM reached 7.6 and 4.2%, the incidence 19 and 11%, and the intensity 40 and 38%, respectively. In 2019, both the IPM and its components maintained a downward trend for Brazil: IPM (2.2%); H (6%) and A (36%). The disaggregated results for the rural and urban areas also showed reductions, both for the IPM, and for (H) and (A). However, the proportions, as well as the falls, were more intense for the rural environment, establishing the regionalist character of the country. The fall in rural IPM was 9 percentage points (p.p.) and urban 3 p.p. between 2005 and 2015. In 2019, rural and urban areas had an IPM of 12 and 1%, respectively. The breakdown by states showed the persistence of the highest levels of poverty in the North and Northeast regions of the country, as well as the perpetuation of the high rate of poverty in rural areas compared to urban areas. It was also found that the greatest deprivation suffered by the population refers to the indicators of the dimension health and basic services, and the smallest, the dimension of housing conditions. As for the comparison between multidimensional and one-dimensional poor, there was a reduction in the proportion of poor in both approaches in the period 2005 - 2015. Especially in the income ranges of R$ 154.00 and U $ 5.50. However, in 2019 the proportion of one-dimensional poor people rose, especially in rural areas. In the analysis of the probability of poverty, it was found that the greatest chances belong to women (unidimensional aspect) and non - white men, living in the Northeast region and who are in a situation of unemployment. While the lowest chances of poverty are those with higher levels of education, living in the urban area and in the South and Midwest regions of the country, and who have a spouse.