Operational Poverty Targeting In Peru – Proxy Means Testing With Non-Income Indicators

Abstract

The measurement of per capita daily expenditures relative to a monetary poverty line, also known as ‘sophisticated means testing’, is the most widely used approach to poverty assessment. However, it is reliant on the implementation of time- and cost-intensive household surveys. Hence, for operational purposes, it is not an effective method for targeting poor households with development services. This paper shows how to identify an alternative poverty assessment tool for Peru. The tool consists of a maximum of 15 indicators that are powerful predictors of per capita household expenditures. The indicators were selected out of a wide range of indicators used to gauge different poverty dimensions. The resultant poverty classification of households is based on the ‘percent point function’ of the predicted expenditures and validated by various accuracy measures and their confidence intervals. The results reveal that the 15 indicators correctly identify over 81 per cent of poor households when the national poverty line is employed as the benchmark. Thus, this tool might be considered, under certain conditions, as an alternative to the collection of detailed expenditure data. It offers an operational instrument for fairly accurate ex-ante poverty targeting and ex-post impact assessments.