Making Aadhaar compulsory jeopardises privacy rights

“I am not aware that this has been thought through” – Jean Drèze speaks to Pragya Tiwari about the new Aadhaar Act

In March, the Aadhaar Bill was passed, empowering the government to use the Unique Identity scheme for delivering subsidies and services and potentially enabling it to require that citizens enroll. However, the bill has come under significant criticism. Pragya Tiwari discusses the context of the latest dispute, and speaks to development economist Jean Drèze about the potential weaknesses of the government’s campaign to link benefits to Aadhaar.

In 2009 the government of India set up the Unique Identification Authority of India (UIDAI) to collect biometric and demographic data on citizens to be stored in a central database and linked to an “Aadhaar” card with a 12 digit unique identity number. In March 2016, 6 years after the world’s largest national identification number project was established, the government passed the Aadhaar (Targeted Delivery of Financial and Other Subsidies, Benefits and Services) Bill, 2016 in the Lok Sobha as a money bill – a move that allowed it to disregard amendments suggested by the Rajya Sabha (upper house), where, unlike in the lower house, the ruling Bharatiya Janta Party does not hold majority of the seats.

The decision to treat it as a money bill has been widely criticised by the opposition and has been challenged in the Supreme Court. The primary objection to the act has been on account of privacy concerns. India does not have codified privacy laws and the act sanctions the collection of a vast amount of data on the citizens of the country without providing meaningful protection against its misuse.

From the inception of the project under the previous government it was deemed voluntary. The Supreme Court in 2013 issued an order reiterating that the use of Aadhaar cannot be made compulsory. The order was a stern warning to a couple of state governments who were insisting on its use for welfare benefits. The act as it has been passed however could lead to Aadhaar becoming compulsory for any subsidy, service or benefit provided by the government. This has further heightened the widespread concern over the privacy of those registered under the scheme.

For most part the debate on the subject has been framed as a trade-off between the risk of breach of privacy of citizens and widespread benefits of using Aadhaar. Nandan Nilekani, the former chairperson of the UIDAI, has been saying that the benefits of Aadhaar will outweigh the privacy risks. The latest economic survey further insists that the use of Aadhaar for welfare benefits along with bank accounts and mobile phones is a ticket to “nirvana.”

However, there is little evidence of these much-touted benefits of using Aadhaar for the disbursal of welfare benefits. While there is no credible impact evaluation study, several scholars are of the view that its potential benefits for welfare schemes might be greatly exaggerated. This is backed by ground reports of the experiences of beneficiaries in states such as Rajasthan and Jharkhand.

Take for example the social pensions made available by the state to senior citizens (aged 60 years or above) living below the poverty line. In the state of Rajasthan the beneficiaries used to receive payment through the post office. The system was prone to leakage, delays and high transaction costs but the move to Aadhaar linked direct bank transfers (DBT) that was supposed to fix these problems seems to have made matters worse.

Iris scanning during Aadhaar enrolment.

Credit: Sonali Campion

The new policy required substantial and rapid organisational and technological change. Beneficiaries were made to open a bank account (roughly 30% of the rural Indians have a bank account currently) and register for biometrically linked unique IDs (UID) that is then linked to their account. Pensions are now deposited directly into these zero balance bank accounts. However, only 27% of villages in India have a bank within 5 km. Therefore, banks have appointed contracted middlemen called business correspondents (BC) who physically carry the money to villages. A point of sale (POS) machine uses UID to verify the identity of beneficiaries via fingerprint authentication.

In practice, the switch over has resulted in several bona fide claimants’ names being randomly struck off from the list of beneficiaries. For those who are still on the list, transaction costs have gone up as sometimes the internet connection does not work and at others their fingerprints are not authenticated, mostly due to cuts and calluses as a result of years of manual labour. Besides, there is no effective way to check against corruption in the new system.

"This would lead to a quantum jump in the surveillance capabilities of the Indian state – a state that has already shown, from to time, a worrying propensity to suppress dissent and curtail the freedom of expression." JeanDrèze

In light of recent events, I spoke to development economist Jean Drèze who has been studying both the social pension scheme and the implementation of Aadhaar, about what the privacy concerns are being weighed against in the name of benefits – for social pensions in particular and welfare benefits in general. Here are the excerpts from an edited interview.

State after state is moving to disbursing social pensions through DBT linked with Aadhaar. What are the ostensible reasons for moving away from existing systems like the use of post offices and panchayat systems?

I am not aware that this has been thought through. It seems to be part of the general drive to make Aadhaar compulsory for as many public services as possible, especially cash payments. There is no obvious reason why Aadhaar enabled bank payments of social pensions are better than other systems such as post-office payments or for that matter Aadhaar-free bank payments.

The term “DBT” is not well defined. In practice, it seems to refer to any Aadhaar-enabled payment. In the case of social pensions, this would require a transition from cash or post-office payments to bank payments. That is not always advisable. In states with a relatively well-developed banking system, there may be a case for this transition. Indeed, some states are already making good use of the banking system for pension payments. But that is not DBT, it is banking.

Where pensions are paid through banks, adding Aadhaar makes no difference since there is very little identity fraud in the banking system. Tagging Aadhaar to a bank account only helps in facilitating the use of banking correspondents (BC).

In areas where pensions are paid through banks, giving people the option of collecting their pension from a local BC is not necessarily a bad thing. But it must be an option, not a compulsion.

Making BC payments compulsory would be very unfair to pensioners, because that system requires many unreliable technologies to work at the same time – biometrics, mobile phones, the internet, and so on. It is simply not an appropriate technology for much of rural India, and that is one reason why the experience with BCs so far has been disappointing.

Are the existing systems uniformly ineffective? Or might there be some that work better than others?

The main options are cash payments, post-office payments and bank payments, with or without Aadhaar. Each has its strengths and weaknesses. For instance, in terms of accessibility, timeliness, convenience, and protection from corruption. Banks are best to avoid corruption, but post offices tend to be more convenient for the recipients, and cash payments are often more timely. The best system I have seen so far is that of Odisha, where pensions are paid in cash on the 15th of each month at the panchayat bhawan (village secretariat).

But this does not mean that other states should adopt the same system. Some may be able to do quite well with banks, post offices or BC, depending on their own circumstances. Aadhaar imposes a one-size-fits-all model that is not demonstrably superior to other ones.

Is it possible to fix the older systems and might that be more cost-effective than moving to DBT?

There is no evidence that the other systems are less cost-effective than DBT. It may well be the other way round, considering all the hassles people are facing in the DBT system, or at least some versions of it. Having said this, much can certainly be done to improve existing systems. For instance, computerisation of post offices would help to achieve greater speed and transparency in post-office payments. Similarly, rural banks need more staff to cope with the rising volume of cash payments under various schemes. Last but not least, whatever the system, closer monitoring is required to avoid delays in pension payments.

Has there been any cost benefit analysis of the use of Aadhaar for the disbursal of social pensions or other benefits?

Not to my knowledge. In fact, I am not aware that any of Aadhaar’s applications have been the subject of a serious cost-benefit analysis so far. Instead, the Unique Identification Authority of India (UIDAI) has tended to rely on public relations, sponsored studies and creative estimates of the alleged benefits of Aadhaar.

The move to Aadhaar linked DBT is seen as desirable because of the central government’s focus on JAM and its belief in the larger promises of this system. You remain skeptical about JAM, in particular Aadhaar. What might some of your concerns be?

Aadhaar, DBT and JAM call for informed debate. Even if Aadhaar is deemed inevitable, there are many ways of using it, some more helpful or harmful than others. In the context of social pensions, the main focus should be on the interests and wellbeing of the pensioners, not on promoting Aadhaar for its own sake.

Coming to the larger concerns about Aadhaar, the main danger is that it creates a powerful infrastructure of surveillance and control. If the champions of Aadhaar have their way, Aadhaar will become an “ubiquitous” means of identity authentication, as Mr. Nilekani himself puts it. The Aadhaar Act allows the government to make Aadhaar authentication compulsory for salary payments, old-age pensions, school enrolment, train bookings, marriage certificates, getting a driving license, buying a SIM card, using a cybercafé – virtually anything. Judging from the experience of the last few years, the government is likely to exercise these powers with abandon and extend Aadhaar’s grip to ever more imaginative domains.

Without invoking any special powers, the government will be able to access most of these databases and link them seamlessly to each other using Aadhaar. This would lead to a quantum jump in the surveillance capabilities of the Indian state – a state that has already shown, from to time, a worrying propensity to suppress dissent and curtail the freedom of expression. Any libertarian, left or right, is bound to be concerned about these developments.

The concern is not just that this infrastructure of surveillance may be misused, but that its very existence stifles the right to dissent. The right to dissent is a fiction without the right to privacy. Aadhaar is an unprecedented invasion of the right to privacy, which is very weak in India in the first place.

With the new bill passed, what safeguards or legal challenges should be pursued?

Two legal challenges have already been initiated. One questions the compatibility of Aadhaar with the right to privacy – the Supreme Court is yet to take a view on this. The other challenges the validity of sidelining the Rajya Sabha by construing the Aadhaar Bill as a Money Bill.

Aside from this, I believe that the best safeguard would be a legal obligation to design Aadhaar as a voluntary facility. This means that alternative means of identity authentication should always be available. This legal obligation would do nothing more than hold the UIDAI accountable to the repeated claim that Aadhaar is a voluntary facility. A series of Supreme Court orders wisely state that if

Aadhaar is voluntary, it cannot be made compulsory for any services to which people are otherwise entitled as a matter of right. These orders, unfortunately, are now overridden by the Aadhaar Act, which gives the government sweeping powers to make Aadhaar compulsory for almost anything. A minor amendment of Section 7 of the Act could undo the damage, but this is unlikely to happen in a hurry.

Coming back to social pensions, while some states are offering no choice at all to beneficiaries in moving to Aadhaar linked DBTs, others are more open to allowing multiple channels of disbursal simultaneously. Is this approach feasible in the long run? Can it ameliorate some of the concerns?

I think so. Workers employed under the National Rural Employment Guarantee Act (NREGA) can choose between bank and post-office payments. There is no reason why a similar arrangement cannot be made for pensioners. Similarly, as I said, BC can be useful as an additional payment option within the banking system. What is dangerous is to impose a single mode of payment that is simply not reliable in much of rural India.

Please note that this blog originally appeared on the South Asia at LSE blog, read the original here.