|Q&A: NANDAN NILEKANI
|Business Standard / New Delhi September 14, 2009, 0:55 IST
|There are concerns on technology, cost and privacy in the decision to allot a unique identification number to every Indian. In a talk with Karan Thapar on the CNN-IBN television channel’s Devil’s Advocate programme, NANDAN NILEKANI, who has agreed to head the newly-created Authority to plan and implement this project, concedes these are legitimate concerns. And, that these can be addressed and the project is worthwhile. Edited excerpts:
Eighty per cent of Indians have Election Commission identity cards, others have ration cards, some people have BPL cards, others have driving licences and passports, there are even PAN cards. Why on top of this do we need a unique identification number?
We need one single, non-duplicate way of identifying a person and we need a mechanism by which we can authenticate that online anywhere, because that can have huge benefits and impact on public services and also on making the poor more inclusive in what is happening in India today.
In addition to name, age, sex, date of birth and address, you actually have the biometrics which are unique to that individual?
Absolutely. It is a combination of, most probably, fingerprints and picture and a biometrics committee will finalise that, but finally that makes it unique. And we will make sure there are no duplicates.
The London School of Economics (LSE) did an analysis of a similar project being considered by the British government and this is their conclusion: “The technology envisioned for this scheme is, to a large extent, untested and unreliable. No scheme on this scale has been undertaken anywhere in the world. Smaller and less ambitious systems have encountered substantial technological and operational problems that are likely to be amplified in a largescale national system.” IIf that is true of Britain, it has to be true of India in spades.
There is no question that we are going into uncharted territories, the technological challenges are immense and one of the risks is the technology.
Not just uncharted territory, this could end up being a case of India’s ambition outstripping its ability. Even today, we can’t issue identity cards with a guarantee that the name is correct or the address isn’t misspelt. We could end by making a complete hash of biometric details.
There are risks but, given the enormous opportunity and developmental benefits it can give, it’s worth taking on so that we get the outcomes we want.
You accept the technology is not just uncharted but not actually fully known?
There is no other country where a billion peoples’ biometrics have been captured and stored in an online database. We don’t have to invent the technology; we have to scale up the existing technology to work at this scale.
The second problem inherent is cost. Once again, the LSE did an analysis of a similar project the British government was thinking of and that is a country one-twentieth the size of India. The LSE concluded the probable cost for Britain would be between 10 and 20 billion pounds. Frontline magazine believes the government in India has a guesstimate of somewhere around Rs 1.5 lakh crore. Is it worth it at that cost?
I don’t know what the exact figure is, but it is much less than that by a factor of 10.
If you don’t know the exact figure, how can you say it is lesser by a factor of 10?
The bulk part is certainly going to be lesser than that.
But it’s a guess?
An informed and educated guess.
So, we don’t know what the exact cost will be?
We don’t know, but I am very confident that whatever the cost, the social, economic and efficiency benefits would make it well worth it.
India is a poor country. This order of money could be better spent if you expand education, health and sanitation, or if you use it to feed the 40 per cent of Indian children who are chronically malnourished.
We don’t want to take away money from important social programmes. But, as we expand our social programmes, their efficiency depends on their reaching the right people and that there are no duplicates taking away the benefits. You need the infrastructure at the bottom to make that happen.
You can only target better those actually availing the benefits but not receiving these fully. Take BPL. The real problem is not leakage, but that there is a vast number who qualify and are not included in the BPL threshold at all. How will you be addressing the second problem?
Today, in a particular state, there may be more BPL cards than the population of the state, because there are multiple cards issued to an individual. With the UID, you will be able to actually trim that down to one card per individual and therefore we will actually know who is not getting this now.
But you can’t identify those who should have BPL cards and do not because they are outside the system, they have been ignored. Technology won’t improve that.
This (UID) is not a panacea for all the problems. This is an enabler which will allow more effective public delivery.
Which is why the order of money involved could be better spent in targeting education, sanitation and health, not to mention child malnutrition, because you would actually then get real benefits rather than what I am describing as notional benefits.
In a country where we are spending Rs 1,00,000-2,00,000 crore a year on different kinds of subsidies and social benefits, to make investment which is a part of that, one-time, to make those investments more efficient, is definitely well worth it.
Is it a one-time investment? Frontline magazine says the government’s estimate of Rs 1.5 lakh crore does not include recurring cost. And we don’t know by how much.
On the scale of money that we spend on public programmes and the ability of the project to deliver better public programmes, it will be well worth it.
I put it to you again, there are so many imponderables about technology, size and cost, that is it wise for a poor country like ours, where there are huge levels of poverty (the Arjun Sen Gupta Committee report says 80 per cent of India live under Rs 20 a day), to be spending this sort of money on this project?
The government has come to the conclusion that this project is strategic and worth it. I have been invited to lead this project. I believe it is viable and I will do my best to make it viable.
How can you ensure the database you are creating will be secure, that it won’t be misused and won’t result in an invasion of privacy?
A very legitimate concern. We are looking at how to make it secure. We are saying nobody can read this database. All they can do is verify the authenticity of an identity. You can ask a question like, is X, X? and the only answer we will give is yes or no. But there is no question that once the UID is implemented and becomes ubiquitous in many applications, then there are challenges of privacy. And, with this project, we have to put in other checks and balances, including laws.
Professor Ian Angle of the LSE, a world renowned authority on precisely the creation of such a database, says with relevance to England, and it will apply even more to India, that what you are going to end up with is the “Olympic games of hacking”. You are going to provide people the biggest challenge to hack through. No one believes in the perfectability of computers, so hackers will hack and succeed.
A legitimate concern and we will have to design it as good as possible. The important thing is — is the risk of hacking and privacy large enough not to do this project? And the view is that the project has so many significant benefits for the poor, in making it inclusive and in giving them a chance to participate in the country’s progress, that it is worth it and we have to mitigate those risks.
You are creating a system which, in the wrong hands, would be a powerful tool for either religious or caste profiling. How can you ensure unscrupulous politicians won’t misuse it?
We are not keeping any profiling attributes in our database. No details of people’s caste?
No. In which case, how can you say to me that you will better target benefits at BPL and other categories? If you don’t know someone is SC or ST, if you don’t know they are OBC, how can you ensure better targetting?
That is the responsibility of the applicant that provides those services.
So, then they will add in that feature into your detail?
That is outside our system. Our system has only basic attributes like the name, address, date of birth. You are creating a weapon which you may not misuse but others could?
Today, we have electronic databases in the country which potentially can be used the way you are suggesting. We are not doing something different from what already exists.
In the UK, the US and in Australia, because the authorities couldn’t respond to public concerns about misuse, they have effectively put on the backburner consideration of similar schemes. If developed countries cannot tackle misuse, how can India, where 35 per cent of the people are illiterate and 22 per cent live below the poverty line?
What these developed countries have put on hold is giving national ID cards to people. But both the US and UK, have a number. In the US, you have the social security number; in the UK, there is the national insurance number. They already have a numbering system, which is what we are going to propose.
Except that it is nowhere near as extensive or as complete in terms of the biometeric details as what you are proposing in India. The national insurance in Britain has been around and developing slowly but it doesn’t have any details that could lead to an invasion of privacy. It doesn’t have any details that can be misused for profiling. Yours could have both.
These are legitimate concerns and we have to address them. But the social benefit, the inclusivity, this project will provide for the 700 million people in this country who are outside the system is immense enough to justify doing this project.
How will you handle the inevitable problems of internal migration or illegal immigration? How will you ensure the wrong people aren’t captured in your system and given an identity and made Indian?
Having this number does not confer any rights, benefits or any entitlements. All it does is confirm that X is X.
There are 100 ways of doing that. Why are we spending close to Rs 1.5 lakh crore just to be able to claim X is X?
To have a system which uses a unique identifier like biometrics, having a system which ensures there are no duplicates and having a system that provides online authentication is, we believe, something that can have a lot of social benefits for the poor.
The LSE conclusion, when they reviewed a potential British concept along the lines of what you are doing in India, was: “The success of a national identity system depends on a sensitive cautious and cooperative approach involving all key stakeholders, including an independent and rolling assessment and regular review of management practices”, and the LSE concluded that did not exist in the UK. If it does not exist there, that environment certainly doesn’t exist in India.
We are trying to make sure all the checks and balances are there. We will have a very wide consultative process. We will involve everybody. We will make it public. All these are legitimate concerns and we have an obligation to meet these concerns