Texto para Discussão, Instituto de Pesquisa Econômica Aplicada (IPEA) 1363
The panel data of the Brazilian Monthly Employment Survey - Pesquisa Mensal de Emprego (PME/IBGE) is actually one of the richest datasets for analysis of labor market in Brazil. The reason is its longitudinal design. Since it is only addressed to labor market investigations, its questionnaire fields only information related to labor income though, not considering other types of income sources, such as pensions, retirement benefits, social transfers etc. Given such a limitation, this paper presents a proposal to improve PME`s utility. This proposal consists of imputing non-labor income at PME, based on an equation system estimated by means of the Brazilian Household Survey - Pesquisa Nacional por Amostra de Domicílios (Pnad/IBGE). Besides describing this imputation model and its consistency in reproducing the same statistics at both PME and Pnad, the paper presents some stylized facts on the dynamics of poverty and inequality in Brazilian Metropolitan Regions (MRs). Due to their dynamic feature, it is indeed impossible to estimate these facts using another Brazilian data that not the imputed PME. Our results basically regard the following issues: seasonality and tendency of poverty and inequality; rates of transition into and out of poverty; and comparison between the incidence rates of chronic poverty and observed poverty.