The Centre has decided, not for the first time, that all Indians should have a Unique Identification Number. The idea itself has been floating around for a few years, with even a few pilots launched here and there, mostly to no avail. This time, however, unlike in the past the government appears to be serious, and has appointed one of the leading advocates of the plan, former Infosys CEO Nandan Nilekani to head the Unique Identification Authority of India, created earlier this year.

Nandan has thought of the need for a national ID system for as long as anyone else, and given his combination of professional expertise and personal interest in this exercise, he is a powerful and practical choice to take up this task. And in giving him this role, the government too has acted well, preferring the specific technology-driven objective of the UIDAI to some other responsibilities. As a member of the Planning Commission or at the helm of the Ministry of Human Resource Development, just to cite two other jobs linked to his name, the difficulties would have been greater (the PC is losing its relevance, and MHRD is a regulatory mine-field better left to a career politician than a techno-manager).

The corporate world has been particularly hearty in cheering Nilekani's appointment, seeing this as new direction in which 'skill' and 'management techniques' may at last be starting to be welcomed in governance too. I think it is hasty to read any such intent for the moment; instead we must see this appointment more as a continuation of an existing trend, analagous to (but of lesser significance) than the earlier selection of Montek Singh Ahluwalia and the Prime Minister himself for their offices. Nandan may simply be the 'man for the job' within the terms of the political walls that now exist.

The 'need' for an 'identifier'

The most powerful arguments for the Multi-purpose National Identity Card (MNIC) have relied on the principle of necessity. Somewhat simplistically, there is a tendency to think that 'developed' economies are those with robust identification systems, and peg the national ID to this kind of thinking. Secondly, it is argued that ID systems are needed to curb the excessive waste and pilferage that is now standard in all our expenditure programs (the UIDAI has an explicit focus on curbing the losses in vast Centrally-funded programs). But there are good reasons not to get carried away by this rhetoric.

  • There are some very developed countries that do not use a national ID card - Wikipedia currently lists Australia, Denmark, Ireland, Japan and the UK among countries without a national identification scheme. Australia recently even shelved its plans to create one, in the face of opposition. And the American Social Security Number, which is the most commonly cited example of the kind of system we 'need', is not really a system for the 'identification' of individuals, although it is used in that way by certain sections of American society (see box).

  • Cleaning up big-ticket Central schemes is clearly a welcome goal. But there is no need to mix this up with the focus of the national ID program. The merit of an objective is not the same as the merit of the effort/method to achieve it. Failures in the delivery of public service delivery may be tackled partially by the creation of identity systems, but this is not necessarily the only or best way to do it. Many services already have their own IDs, in fact - this is the clearest evidence that their poor performance is not due to a lack of ID systems or technology, but a broader failure of governance.

  • Linking the ID to benefits under any of the ongoing programs is a risk. Practically speaking, the benefits cannot be linked to the ID itself, since (to begin with, and for some years) most of the beneficiaries of various programs will not have completed the identification process. This means, inevitably, that we will have a certain period during which benefits will have to be delivered to those without IDs, or denied to those unable to establish IDs at the point of service. Indeed, these are problems that already exist for most programs, and saddling the UIDAI with resolving these is a mistake.

Let us also not forget another important reality. An ID only makes sense if it reveals something about the person whose identity it contains. Usually, this will include some personal details (age, sex, date of birth, caste, religion, and so on - some of these are particularly needed for programs where benefits are linked to one or more of these pieces of data) but it is not not clear that people want to be identified clearly in all cases. In India, identity is already a potent issue, and the cause of endless ethnic and religious conflict - a system to improve identification, therefore, will inevitably raise concerns over potential abuse. And our politics, which is essentially defensive, isn't ready to deal with this.


In the US, Social Security numbers were introduced for a pension program based on employer contributions, and were not intended for broad ID purposes. And despite this limited focus, it took five years for the program to get off the ground. Even today many agencies of the US government and the states do not use it to establish identity at the point of delivering services or benefits


 •  E isn't everything
 •  Interview with Nandan Nilekani
 •  Book Review: Imagining India

An additional difficulty, in the case of the UIDAI, is the expectation that it should deliver something within two or three years. This may be too much to expect, given that much of the 'implementation' would need to negotiate its way past the countless array of programs that the Centre and the States have, each with its own well-guarded administrative and political turf. In the optimism over Nandan's appointment to the UIDAI, we should not forget the complexity of the challenge ahead.

Fixing the goal posts

Given all this, what are the options before the UIDAI? The Government has already put its foot into its mouth, by linking the Authority to the specific goal of reducing irregularities in its flagship social welfare programmes. Therefore, this is not a focus the UIDAI can withdraw from. The best alternative would be to pursue a few other objectives which are critical, and hope that favourable outcomes on these other fronts can be leveraged to overcome obstacles to the stated goal of cleaning up Bharat Nirman or the NREGS.

  • Distribute, distribute, distribute!
    First, the Authority must decouple the goal of getting people identified from the question of how these IDs will be used. The distribution of IDs is an idependent goal, and should not be linked to questions about where they will be used, or even what they will reveal about those being identified. If this view is embraced, then a program of rapid ID distribution can be taken up.

    Anyone who wants an ID should be given one with minimum fuss, as long as any other government-issued ID is provided as documentary evidence. The Authority should launch an all-out effort to distribute as many IDs as possible through as many channels as available, so that numerous touch-points for getting hold of an ID are accessible to citizens. Once IDs are issued, it can embrace the techniques of the internet era, allowing citizens to 'crowd-source in' their documents to link these to the IDs they are assigned. Large institutions, with tens of thousands of employees, can also be roped in to facilitate distribution of IDs to their people, in controlled environments. It is also likely that as large government organisations are roped in to faciliate, some of them are likely to figure out ways to start linking their own employee and benefits documentation to these IDs, for their own effectiveness.

    Still, and inevitably, this will raise some cynicism that the ID is a Trishanku object, floating around in a universe of non-use. This has to be addressed, but it must not be allowed to derail the distribution of the IDs themselves.

  • Acceptable proof of identity
    Secondly, the UIDAI must launch a parallel effort to get as many institutions as possible to accept the identification it issues as proof of one's identity. There is no shortage of use cases for this - post offices, banks, the tens of thousands of security check points everywhere in the country, trains and airlines, and many others should be encouraged to accept the Unique ID as a substitute for the current forms of identification they now accept. Campaigns of advertising encouraging people to 'leave your PAN card safe at home, use UID to prove who you are' should be all over the place. Indeed, the Rs.100 crores set aside for this project is most needed to communicate the program, rather than actually initiate it.

  • Linking with schemes and programs
    Whatever the gains on the above two counts, the UIDAI will have to deliver at least partially on its stated goal of making government programs work well. Here, what the Authority needs more than anything else is a working arrangement with a few states that ensures implementation of the proposed vision. A tightly coupled program which ensures that the IDs, once issued by the Authority, are seamlessly introduced into a few government schemes at the state-level, is needed. Karnataka is a good bet, since Nilekani himself heads the e-governance effort there, and Karnataka is already among the leading states when it comes to computerised administration.

The National ID certainly has the potential to transform the delivery of public services in important ways. But this is not necessarily its biggest benefit. The real value of such programs is to create a backbone upon which governance and economic development rest comfortably. The ID should be a catalyst for such change, rather than merely a tool for authentication or auditing government schemes. This is all the more important because there has been no proper debate on the need for the ID scheme, and consequently no preparation for the very real pitfalls that lie ahead.