Many social cash transfer programmes in poor countries are targeted on the poorest people through a "proxy-means test." This is a way of estimating the hard-to-measure actual poverty of a household through an easy-to-measure proxy - the ownership of assets such as quality housing or a vehicle.
An alternative method of targeting is "self-targeting" in which the experience of receiving assistance is made unpleasant to discourage the rich from applying - such as through requiring manual labour.
A new paper adds a simple layer of self-targeting onto the existing proxy-means test for a programme in Indonesia with good results. Instead of travelling round to households to administer the test at their home - participants are simply requested to travel themselves to an office to make their application. This simple added inconvenience made a big difference to the effectiveness of the targeting.
An alternative method of targeting is "self-targeting" in which the experience of receiving assistance is made unpleasant to discourage the rich from applying - such as through requiring manual labour.
A new paper adds a simple layer of self-targeting onto the existing proxy-means test for a programme in Indonesia with good results. Instead of travelling round to households to administer the test at their home - participants are simply requested to travel themselves to an office to make their application. This simple added inconvenience made a big difference to the effectiveness of the targeting.
per-capita consumption was 13 percent lower for beneficiaries in the self-targeting villages than those under the status quo. Moreover, exclusion error was actually less of a problem in self targeting than in the status quo: the very poorest households were twice as likely to receive benefits in self-targeting than in control areas.Alatas, Banerjee, Hanna, Olken, Purnamasari, and Wai-Poi, Ordeal Mechanisms in Targeting: Theory and Evidence from a Field Experiment in Indonesia, http://economics.mit.edu/files/8449
4 comments:
One has to be careful here though - note that they find that once the decision to move to self-targeting has been made, experimentally increasing the distance to the testing centre doesn't actually improve targeting, but does reduce the number of applications.
This leads them to the difficult-to-parse conclusion that imposing an "ordeal" improves targeting, but making current ordeals more unpleasant won't. Makes you wonder how easy it is to get the unpleasantness of the ordeal right in the first place.
True. I think ordeal is a bit of a strong word in this case, and there are lots of other benefits to establishing an office where people can come to apply, and ask questions about available benefits - in theory this could allow for continual applications rather than an ad-hoc periodic retargeting.
Does this form of targeting basically eliminate the disorganized?
Yep
Post a Comment