The government of Sri Lanka has increased spending on the leading welfare scheme Samurdhi, from LKR 15 billion in 2014, to LKR 43 billion in 2016, almost a tripling. There are 1.4 million beneficiaries, classified into those can be graduated out of the scheme, those who cannot, and those in between. Apparently, another 1 million people are clamoring to be included in the scheme. In this seven-minute speech made in a Parliament recently, Dr Harsha de Silva provides a quick comparison of the three principal methods of ascertaining poverty.
We know, from our experience with teleuse@BOP surveys, that getting an accurate fix on levels of income and assets (poverty) is not easy. An experiment in Indonesia suggests that asking the village to decide is superior, because it is almost as accurate as the standard method which uses household assets and generates greater buy-in (less complaints). One thing I did not get from the writeup was relative cost. Convening a community meeting is not costless. Neither is the asset-based method.