Why poor need Unique ID
Government programmes are open to misuse by people who run them. The unique identity card project can fix the problem if it is backed by political will, report Abhijit Banerjee and Sriram Raghavan.india Updated: Nov 09, 2009 22:56 IST
One of the biggest challenges India faces today is that of delivering more effectively to the poor. For the last many years, the below poverty line (BPL) card has been the primary basis of all redistributive programmes in India. The trouble is that the cards often seem to end up in the wrong hands.
A study conducted by the Planning Commission concludes there are 23 million more ration cards than households, and their guess is that most of these are BPL cards. The study also emphasises mis-targeting. It estimates that in all the major states save four, more than 40 per cent of households have the wrong kind of card [BPL households with above poverty line (APL) cards and the other way round].
Moreover it is clear that a substantial part of these “errors” are deliberate: A detailed study of 173 villages in Raichur district in Karnataka by Atanassova, Bertrand and Mullainathan, finds that about one-third of the eligible households, based on the official criteria, don't have a BPL card, while about half of the ineligible households do. More worryingly, when they use income as a proxy for wealth, the poorest among all ineligible households are not the ones who have the card, which is what one would expect if this was an innocent mistake. Being socially connected to village officials turned out to be an excellent predictor of where these mistakes were concentrated.
In addition to faking and mis-targeting of cards, the Planning Commission report suggests that there is a lot of capture — BPL cards issued to a BPL family that end up in the hands of someone who is not BPL; grains that are supposed to have been sold to a BPL family that are actually sold on the open market, etc. The result of all these problems, the report concludes, is that in the case of the targeted Public Distribution System only 42 per cent of the grains intended for the poor actually reach them. Similarly, a recent study of the National Rural Employment Guarantee Scheme (NREGS) by Niehaus and Sukhantkar from Harvard University, based on data they collected from about 1,500 households in Orissa, suggests massive capture — when the official NREGS wage goes up to Rs 70 a day, workers continued to be paid Rs 55 (moreover, only 40 per cent of the reported workdays seemed to be real).
This sort of malfeasance in government programmes, of course, goes beyond anti-poverty programmes. There are news reports of thousands of fake employees in government departments. A recent survey revealed that nurses in the government health system in Udaipur district show up for work less than 40 per cent of the time.
Benefits of UID
The net result of all this is that the effectiveness of public expenditures is very low, and the poor who depend most on public assistance are the ones who pay for it directly. It also undermines political support for public expenditures, which can rebound on the ability of the government to sustain its anti-poverty efforts.
A unique ID (UID), interpreted as a data-base that matches each person’s biometric identifiers with a name and a number, and a system by which it is possible to check that a person who claims to be identified with a particular number is indeed that person, will help in a number of ways. On the issue of fake identities, the advantage is obvious: ghosts do not leave finger-prints. If you can pay only people who have a UID, you can pay only real people, and since the identity is unique the same person cannot be paid twice.
A second, perhaps less obvious, benefit is with respect to mis-targeting. Since all systems will use the same UID, it would be easy to link up various databases. One would not need special surveys to tell us that someone who has a BPL card also got a loan of Rs 10 lakh from a government bank — the right command given to computer will instantly list such people who could then be automatically removed from the BPL list.
Making delivery effective
A third potential benefit is with respect to capture. The owner of the fair-price shop, who sells the grains on the open market, will at least need to have the co-operation of the person against whose UID the grains are being issued, as long as it is required that the identity is established before the grains are handed over.
Finally, the ability to identify the presence of a specific individual makes it much easier to centrally monitor delinquency among government servants. In principle, this could be used to make sure that people actually come to work.
Of course, all of this will require political will and other, more institutional, changes. Technology can only enable — it cannot make anything happen. But it is clear that this has the potential to shift the conversation about delivering more effectively to the poor.
(Banerjee is Professor of Economics, MIT; Raghavan is chief executive officer, Comat Technologies Private Ltd)