Targeting Assistance: Electricity Consumption is a Superior Method

This is one of several indicators previously identified by the Sri Lankan government, through a gazette notification in 2019, as potential components of a proxy means test to identify low-income families for welfare benefit payments.15 In the case of Sri Lanka, 99% of households are connected to the national grid, and 48% of the population lives in households that consume 60kWh or less of electric. [...] For the poorest 10% of the population, i.e., the most vulnerable group, targeting via the electricity use criterion captures 81% of households whereas the Samurdhi system only captures 40%. [...] We reach the same conclusion by examining the poorest districts in the Island, as measured by the number of people in poverty.17 Four of the five districts with the largest number of poor people are among the five districts with the most people living in households with low electricity use. [...] Calculations from household survey data indicate that 74% of the poorest 60% of the population will receive benefits according to the electricity use criterion (with an exclusion error of only 26%), whereas only 32% of the same group receive Samurdhi benefits - with an exclusion error of 68%.18 Exhibit 3: Percentage of each decile that would receive cash transfers by using criterion of electricity. [...] If coverage of the poorest decile, and non-coverage of the richest decile are the only objectives, and are given equal weight, the targeting mechanism of household use of 60kWh or less is the best to use.