| Abstract: | Geographic targeting is perhaps the most popular mechanism used to direct social programs to the poor in Latin America. This paper empirically compares geographic targeting indicators available in Peru. To this effect, I combine household–level information from the 1997 Peru Living Standards Measurement Survey (LSMS) and district–level information from the 1993 Peru Population and Housing Census. I then conduct a series of simulations which estimate leakage rates, concentration curves, the impact of transfers on poverty as measured by the headcount index, poverty gap and P2 measures of the FGT family, and non–parametric (kernel) densities when transfers are based on alternative indicators. I conclude that there is substantial potential for geographic targeting in Peru. However, the differences in outcomes across geographic targeting indicators are small, and are not statistically significant. These results are in keeping with earlier work which suggests that (among reasonable alternatives) the choice of geographic targeting indicator does not have an important bearing on poverty outcomes, and are at odds with more recent research which stresses the advantage of poverty maps which "impute" consumption or income. |
