How best to target the poor: an operational targeting of the poor using indicator-based proxy means tests

Households whose incomes are below the poverty line are considered eligible for targeted benefits. The current paper seeks to find out how best to target the poor, using household survey data from Uganda. The study compares the performances of four alternative models and analyse the sensitivity of the models to the chosen poverty line.

The authors reveal that there is no single national poverty line in Uganda. Instead, the poverty line is disaggregated into different regional poverty lines, which reflect the differences in costs of living between the four divisions of the country.

Main findings are:

• under the national poverty line, the poverty rate is almost the same as the rate according to the one-dollar international poverty line 
• as poorer households tend to have higher sizes, the poverty rates are higher when expressed in percent of people
• development policies can be very effective in reaching Ugandan’s poor, especially those living below a $2.15 a day poverty line if targeted using the models developed
• none of the applied models can be deemed more target-effective in any particular expenditure quintile
• when using Logit model, potential beneficiaries with the support of community representatives should be allowed to appeal if they think that they qualify for benefits