DIFFERENT IMPACT CHANNELS OF EDUCATION ON POVERTY

This article analyzes both the monetary and non-monetary effects of theeducation level of the head of the household on poverty. We propose thatschooling returns should not be thought as a single number - usually theschooling coeffi cient in an income equation - but as a set of elements whoselength depends on the number of identifi ed poverty dimensions. The monetaryanalysis employs the Quantile Regression technique, very helpful especiallywhen one is interested in extremes of the income distribution function. Ourresults show differences across quantiles of the returns. We also found interestingdissimilarities by gender and urban-rural location. Exploring thenon-pecuniary returns, we found that the education of the head positivelyinfl uences family health and housing conditions.