When Communities Identify Their Own Poor, Aid Has The Most Effect

When governments and NGOs plan on giving assistance to the most needy, how do they know who needs the most assistance? It’s a question people are at great pains to answer, yet social welfare programs around the world are still plagued by error and abuse. That has not deterred the development of programs to help people escape extreme poverty. But because the poor–especially those earning less than $2 or $3 per day–typically hold informal jobs with no official records about their earnings, conventional ways to determine need are ineffective.

MIT researchers in a recent study (PDF) looked at two alternative methods for establishing who needs assistance. For 640 villages in Indonesia, they told the communities to figure out for themselves who their neediest families were, and those people would be given aid. They compared this to a survey of household assets including homes (wall materials, structure types), assets (TVs or motorbikes) and the level of education of the head of household. The results suggest that community selection, rather than screening based on objective measures of household wealth or consumption, was slightly less accurate relative to matching aid to incomes, but provided much greater satisfaction in the community compared to the empirical measurements.

Fairness and poverty, as it turns out, are somewhat subjective. There was only a minor difference in accuracy between the two methods, but the researchers found the community approach led to 60% fewer complaints, and far fewer difficulties distributing funds, compared to objective methods in the villages. And awarding aid according to measured assets proved less effective than the judgment of the community when selections made adjustments for life circumstances such as widowhood, disability, and serious illness.

“Communities appear to apply a different concept of poverty,” said the authors. “The community seems to have a widely shared objective function other than per-capita consumption, and implementing this objective is a source of widespread satisfaction in the community. In exchange for a slightly less effective method of determining who needs the most help, governments can gain significant amounts of credibility by empowering local citizens to help distribute aid.”

Testing this program in other countries on a wider scale would be the next step. In Indonesia, where existing cash assistance programs for the poorest one-third of the population misallocate 45% of funds to non-poor households, while excluding 47% of needy households, according to a World Bank analysis (PDF), it couldn’t come soon enough.