Indian Express, 3rd December 2012
Financial justification for Aadhaar doesn't require it to cover entire population or have multiple uses
Some people think of Aadhaar as a magic bullet for India. Others oppose it for privacy concerns. The government has showcased Aadhaar as a tool for targeted subsidy payments. As with all government programmes, the public should be sceptical, and the government must demonstrate through a cost-benefit analysis that the expenditure of public money is justified. Aadhaar can only address the issue of subsidies having ghost and multiple beneficiaries. In addition to the money spent on Aadhaar, there are the complexities of Aadhaar-integration for each subsidy scheme. The costs are front-loaded, the benefits come much later. Is it worth building Aadhaar? In a recent study at the National Institute of Public Finance Policy (NIPFP), we undertook a cost-benefit analysis of Aadhaar from this perspective. We find that the internal rate of return on building Aadhaar is over 50 per cent. This suggests that we should proceed with Aadhaar in order to run subsidy programmes better. The concerns about privacy can be reduced by limiting Aadhaar to those individuals who benefit from subsidies.
The main conclusion of the study is that it is worth undertaking the expenditure on Aadhaar, if only to plug leakages arising from ghost and duplicate beneficiaries. The financial justification for Aadhaar does not require it to cover the entire population, and it does not require the scheme to have multiple uses.
In developing countries, proposals to spend money on subsidy programmes are generally seen positively. We think that having good intentions in establishing a government programme or a government agency is sufficient for good results. What is needed, instead, is clarity of purpose for each government scheme, activity or agency. Once the objectives are clear, we should be measuring outcomes, and asking about the extent to which the stated objectives were met. The moment outcomes are measured, it becomes possible to ask bang-for-the-buck questions: Is there a way to achieve this goal at a lower cost? Given two different ways to achieve a stated outcome, which one is better?
The subsidy programmes run by the Indian state suffer from immense problems that come from not asking such questions. Once we look beyond the halo of moral purity, there is, typically, a lack of clarity on objectives, failure to deliver on objectives, scanty knowledge of how much money is being spent where and of who the beneficiaries are.
For example, the purpose of the public distribution system (PDS) is to deliver cheap wheat and rice to the poor. It is easy to calculate how much it would cost if, say, 100 kg of cereal per year were gifted to 20 per cent of the poorest people in India. What we have instead is a vast and sprawling enterprise that distorts the market for cereals, which is characterised by theft at various levels, and has failed to eliminate hunger among them.
Many people who think about public policy in India are fairly convinced that a biometric identification system would help reduce leakages in subsidy programmes. But while the expenses of having the Unique Identification Authority of India (UIDAI) and enrolling everyone are evident, there are also substantial integration costs when programmes such as the PDS are UID-enabled. At the same time, the scale of waste in existing subsidies is very large. The UIDAI is not a magic bullet either; it will not solve the problem fully. It will only solve two things: payments to non-existent persons, and payments to one person multiple times.
The key impediment to a high quality cost-benefit analysis of UIDAI is the lack of data and research about existing subsidy programmes. The very problems of subsidy programmes, as presently run by the Indian state, make a precise analysis of improvements in their process engineering difficult. The NIPFP study overcame this constraint by making a series of conservative assumptions.
When these estimates are put together into a formal cost-benefit analysis, they demonstrate that the internal rate of return on building UIDAI is around 50 per cent in real terms. We often find that discussion on costs and benefits turns into a disagreement about assumptions. To allow analysts to modify assumptions, a spreadsheet with the full calculation, clearly showing all assumptions and formulas, has been released on the Web. This makes it easy for anyone to modify the assumptions and get new estimates if they disagree on some of the assumptions.
The construction of the UIDAI, and the consequent transformation of the existing subsidy programmes, is thus well justified. If the government must run subsidy programmes, it should make sure there are no leakages. These leakages are not just about wasted money: we also have to worry about the political economy of business strategies that are rooted in subverting state structures and stealing.
That leaves a distinct policy debate about the problems of privacy. There is merit in civil liberty groups' concerns about the threats to freedom in India, as well as in the concern about the implications of better data in the hands of the government. A reasonable compromise, which could satisfy everyone, consists of emphasising the use of UIDAI for the beneficiaries of subsidy programmes. For the people who wish to take money from the government, we would intrude on their privacy to the extent of their having an Aadhaar number and potentially suffering from the consequences of greater tracking. It should be possible for a person who stands on his own feet - who does not even buy subsidised LPG - to organise his own life with zero tracking by government or security agencies. Such an approach, where one is vigilant about information in the hands of government, would strengthen the foundations of Indian democracy.