Resumen
Chile?s recent success in reducing poverty has been remarkable, with poverty
rates falling by approximately two-thirds over the last 20 years. However, further
gains against poverty may be inhibited by the increasing difficulty associated
with targeting fewer poor people and better, more localized information may thus
be required. Moreover, public policy that seeks to understand the relationship
between poverty, socioeconomic status and urban or rural residence depends on
survey data that may not be representative at low levels of aggregation; although
censuses are representative by definition, they generally lack detailed income data.
Recent mapping techniques combining census and survey data afford detailed
descriptions of the spatial distribution of poverty. Indeed, such techniques enable
the calculation of robust and statistically precise estimates of poverty at low levels
of aggregation, allowing a more nuanced picture of poverty to emerge. In this
paper we use such methodologies to produce consistent estimates of both rural and
urban poverty at the county level in Chile. The results demonstrate considerable
heterogeneity in local poverty, suggesting that future antipoverty policies must
be targeted at the local level. Moreover, the estimates we provide may be used
to further analyze the causes and consequences of localized poverty.