Numbers and NREGA

ON 7 JANUARY 2011, a group of science students staged a silent protest as Nandan Nilekani began delivering a lecture on ‘Adhaar’s role in the transformation of public service’ at the National Institute of Advanced Studies in Bangalore. The students were holding printed posters that read: “Secure electronic archive is a myth.”

Nilekani, a husky man in his mid-50s, left his company, Infosys, in 2009 to take up a new role as the chairman of the Unique Identification Authority of India (UIDAI), a government agency charged with creating an identification number for every resident of India. The project, known as Aadhar—meaning “foundation” or “basis”—aims to create a universal database, backed with biometric data such as fingerprints and retinal scans, capable of verifying the identity of every Indian. The goal, as Nilekani has said, is to establish “one single, non-duplicate way of identifying a person.” The project’s supporters argue that UID will simplify registrations and transactions for rural or poor Indians, eliminate fraud and corruption in the distribution of public funds and goods, and bolster national security against illegal immigration and terrorist threats.

But as the protestors in Bangalore demonstrate, the initiative remains hotly controversial: its cost has been estimated at 1.5 trillion rupees, and it aims to cover a population of 1.3 billion people—likely the largest numbering process in human history. Activists have raised a series of further questions about the programme. Can the security of the central database, containing personal information, be guaranteed? Will this storehouse of personal information be misused by police or intelligence agencies? Will those who fail to enrol—in what has been advertised as a voluntary programme—be excluded from government services or benefits?

In late August 2010, the activists intensified their anti-UID campaign with newspaper op-eds, TV interviews and blog posts outlining their objections to the project after the Ministry of Rural Development signed an agreement with the UIDAI to make possession of a UID number compulsory for participation in the National Rural Employment Guarantee Act (NREGA). The job cards of more than 20 million workers in the five-year-old employment scheme—the main plank of the Congress party’s pro-poor policy—are set to expire in 2011. The fear expressed by the activists, in short, is that the requirement for UID enrolment will prove an obstacle for workers; the numbering scheme, which is still in its early stages, remains untested and unproven.

Five years after its introduction, NREGA has made a decisive impact on the lives of India’s rural poor: the job scheme covers 619 of India’s 626 districts; it has dispensed 784 billion rupees and employed 44.1 million people. Villages that had been idle for decades buzzed with activity: harvesting rainwater, planting trees, digging canals, laying down drains and roads. The programme is not without its problems: studies have found ample evidence of corruption on the part of local pradhans and administrators, who have siphoned off NREGA funds by inflating daily attendance or drawing money for fictional workers.

For Nilekani and the backers of the UID project, the irregularities in NREGA and other large-scale social welfare schemes provide the most compelling rationale for introducing a universal identification database, which they suggest will eliminate fraud and ensure that government funds pass directly to labourers. According to UIDAI, worksites will be equipped with devices for fingerprint capture and authentication in order to prevent the compilation of false attendance records or payments to “ghost workers.” Through handheld devices, the attendance data would be transmitted into databases by using mobile phone or nearest Internet connectivity. It would be harsh to question this idea as it has the potential to speed up the payment of wages. But the law already contains provisions intended to detect fraudulent accounting through “work measurement,” which requires tangible evidence of asset creation. What is needed now is for these rules to be backed with a more robust regime of inspection and enforcement. Though attendance fraud has been given wide attention in the media, by far the largest amount of corruption in NREGA comes from false receipts for the acquisition of construction materials, not the invention of “ghost workers.”

The use of biometric data, according to the UIDAI, will further ‘financial inclusion’ among the rural poor by enabling cash transfers outside of the banking system: if the local corner store is equipped with a device to read fingerprints, a worker or pensioner could verify his identity and receive a payment from the owner of the store, who will in turn be reimbursed by a bank or government agency.

At present, the government of India, as per the muster roll data, deposits the wages into the labourer’s bank account. Contrary to popular assumptions, 83 percent of NREGA job card holders already have bank accounts, and payments in cash are no longer used in most parts of the country. The labourer goes to the nearest bank and stands before the cashier, who after identifying him from his passbook, distributes the money. The cashier cannot refuse money to the labourer; and the labourer cannot deceive the cashier. The same cannot necessarily be said of the store owner.

At the same time, the UIDAI’s own “Biometrics Standards Committee” has noted that retaining biometric efficiency for a database of more than one billion people “has not been adequately analysed” and the problem of fingerprint quality in India “has not been studied in depth.” Here the accuracy of fingerprint matching is the point of concern: the fingers of labourers are prone to cuts and scars while working, which can lead to a negative reading from the biometric device. What if someone’s fingerprints won’t match? The UIDAI has suggested that retinal scans will provide a backup method for identification—but these are expensive, and it would be impossible to conduct daily identification across thousands of worksites using a retinal scanner. How will workers prove their identity if the fingerprint reader rejects them?

It is possible, of course, that a properly functioning UID database and the successful deployment of all the required technology and training could indeed improve the efficiency of NREGA. The sheer size and complexity of the job scheme, which makes it an ideal target for the backers of the UID project—who are eager to enrol as many people as quickly as possible—also makes it unlikely that UID can be seamlessly integrated into NREGA without disrupting the programme and hurting the millions of people in poverty who depend on its wages.

The government’s decision to make UID enrolment mandatory for work under in NREGA runs the distinct risk of limiting participation in the jobs scheme: it is hard to imagine that the 23 million workers whose job cards are set to expire will join UID prior to the deadline. The process of enrolment—which requires the completion of multiple forms and the registration of fingerprints—is not simple, and its details and prospective benefits have not yet been made clear to the rural poor who are supposed to be its primary beneficiaries. Further awareness campaigns on this front are still required. Given this reality, the government’s decision to make the possession of a UID number compulsory for job card renewal may prove dramatically detrimental to NREGA.

The government has batted away inquiries about UID and NREGA with a series of vague responses. Mihir Shah, a member of the Planning Commission, told me that he believes “putting UID into NREGA is the way forward.” He admits that there are likely to be “teething problems” at the outset, but says he believes that in the long run the UID system will be better because “the electronic database will be secure.”

Nilekani did not respond to the protestors at his lecture in Bangalore, and continued to emphasise the positive benefits of UID: “It’s an opportunity for people to open bank accounts, have micro-ATMs and mobile phones,” he said. All this may indeed be true, and yet none of it proves UID is necessary, or beneficial, for the health of NREGA—which should not be used as a registering agency for identification numbers.