Non-labor income imputation in the Brazilian Monthly Employment Survey (PME/IBGE) and its utility in dynamic analysis on poverty and inequality

Abstract

Though the Pesquisa Mensal Emprego (PME/IBGE – Monthly Employment Survey) is the richest dataset for the analysis of labor market dynamics in Brazil, it does not contain information on workers’ non-labor income. This paper proposes a new methodology to impute this income source into PME based on a system of equations that is estimated from the Brazilian Household Survey (PNAD/IBGE). Besides describing the imputation model and its consistency, the paper also presents some stylized facts on the dynamics of poverty and inequality in the main Brazilian metropolitan areas. Our results cover the patterns of seasonality and tendency of poverty and inequality; rates of transition into and out of poverty; and the comparison between the incidence rates of chronic poverty and observed poverty.