“How does Aadhaar establish identity and why is it so important?” queried the journalist. Ram
Sewak Sharma, director general and mission director of UIDAI, said, “This number will be the foundation through which citizens can claim their rights and entitlements from various agencies and services.3 We decided to rename UID [unique identification] as Aadhaar. We wanted a name that had national appeal, could be recognized across India and resonate in different languages, besides being easy to remember.”4 (See Exhibit 2.)
The journalist wasn’t satisfied. “This will be another way to profile or divide us.” A UIDAI deputy director general clarified, “Aadhaar is devoid of classification based on caste, creed, religion and geography . . . It’s a number, not a card; it establishes identity, not citizenship; it’s voluntary, not mandatory.”5 The journalist continued: “There are vested interests who benefit from the current system. How will you ensure the project isn’t derailed?” Nilekani explained: “The most important growth driver is expanding access to resources and opportunity . . . Information technology (IT) has a huge role to play . . . in enabling an open, inclusive and less corrupt society.”6 Nilekani continued:
There are two big things we have to do. We have to constantly focus on enrolling the unenrolled and make sure there are useful applications for the enrolled. Think about it like a flywheel that has to be put into motion so it gains acceleration. The real strategic challenge is how do we create a virtual flywheel, which spins faster and faster, and gains so much momentum that it is unstoppable.
Sharma added, “One cannot improve human beings, but one can improve systems. The same flawed human beings with a better system will be able to produce better results.”
Nilekani smiled at his team. They were so different, yet so aligned. Sharma, with the Indian
Administrative Service (IAS) since 1978, had been the first officer in Bihar to introduce a computer into the workplace. He computerized the treasury department in Purnia district, Bihar’s public grievance system, and the Mahatma Gandhi National Rural Employment Guarantee Act program
(MGNREGAa). Nilekani thought of the volunteers, interns, and people who had joined the team on sabbatical from the best Indian and global companies. Despite their different backgrounds, they were united in their dream to “do something transformational” for India. Nilekani’s thoughts drifted to the tasks needed to establish the largest data management program in the world.
India: Unfulfilled Promise
India, often described in the 21st century as “an emerging superpower,” remained a country of paradoxes. After 1991, India had begun to move away from the “license raj,” with economic liberalization, industrial deregulation, privatization of state-owned enterprises, and reduced controls on foreign trade and investment. By early 2011, the economy was posting robust growth, with a burgeoning middle class.
Yet prosperity was slow to trickle down to the base of the pyramid. Traces of old autarkic policies remained, along with long-term challenges of poverty, insufficient access to quality education, rural to urban migration, and endemic corruption (see Exhibit 1). India had 22 official languages, radio stations in 76 languages, newspapers in 90 languages, and both the second-largest English-speaking a MGNREGA was the GoI’s largest social sector scheme, guaranteeing 100 days of wage-employment in a financial year to a
household, and covered more than 120 million Indians in rural India. (Source: http://nrega.nic.in/netnrega/home.aspx, accessed August 25, 2011.)
2
Aadhaar: India’s ‘Unique Identification’ System
712-412
population and the largest illiterate population in the world. 7 With 50% of its population under the age of 25, India was one of the world’s youngest countries, led by an 80-year-old prime minister.b It was also one of the most populous. Yet 75 million were homeless. Further, although Indians no longer had to wait to buy a car, phone, movie or air ticket, many things were in short supply—rail tickets, admission to schools and colleges, and government jobs. Long waits for basic services, including passports and driving licenses, and repeated visits for common services such as bank accounts, were common. Indians joked that the “shortage mentality” was so prevalent that if Indians saw a line, they would queue up without knowing what was on offer. Once the window for distribution opened, they would break ranks and rush to the front, fearful that by their turn, stocks would have vanished.
A British colony till 1947,c India was the world’s largest democracy half a century later, but was riven by politics based on caste, creed, and coalitions.d The Preamble to the Constitution described
India as a “sovereign socialist secular democratic republic”; concern for the aam aadmi (Hindi for
“common man”) figured in all political manifestos. The Congresse-led United Progressive Alliance
(UPA) government, in power since 2005, broadcast the mantra of “inclusive growth.”f (See Exhibit 4.)
However, it was estimated that “Below the Poverty Line”8 (BPL) families paid $203 million in 2008 to avail themselves of free public services. Reports alleged that in Uttar Pradesh (India’s most populous state), more than 80% of the food and fuel aid distributed through the public distribution system
(PDS) was systemically stolen by a widespread network, which included bureaucrats, transporters, village council leaders, and “fair price” shop owners.9 On August 15, 2011, Anna Hazare (74), an exmilitary man known for his personal austerity, went on a public “fast” demanding the passage of a bill to contain corruption and appoint an independent body termed a Lokayukta (anti-corruption ombudsman organization). The UM aadmi (“upper middle class man”) took to the streets and signed up for e-mail campaigns, demanding public accountability. (See Exhibit 1.)
In 2006, the UPA government approved a project called “Unique ID for BPL families” under the department of information technology (IT). Meanwhile, the Registrar General of India, under the
Citizenship Act of 1955, was to create a National Population Register (NPR) and issue multi-purpose identity cards to Indians. An Empowered Group of Ministers (EGoM) was constituted to collate both schemes. On January 28, 2009, the government set up UIDAI as an attached office to the Planning
Commission, under the Ministry of Planning, headed by the Prime Minister. (India had followed a centralized planning model since 1950.) The Planning Commission formulated plans for effective resource utilization and was led by deputy chairman Montek Singh Ahluwalia, an Oxford-educated economist, whose career included 14 years with the World Bank and IMF as well as a stint as finance secretary to the GoI.10 The initial team was to be 115 officials.11 (The Finance Ministry noted that identity proof to the poor would enable smoother delivery of benefits.)
Some argued that there were cheaper ways to deal with corruption. For example, a state in central
India had used smart cards to reduce fraud in the delivery of subsidized food. Usha Ramanathan, a lawyer focused on civil liberties, criticized the UIDAI approach: “This is a solution in search of a problem.”12 Others maintained that more was required, including effective delivery systems, transparent governance, and monitoring agencies independent of bureaucracy and politicians. 13 b Dr. Manmohan Singh was born in a village, now in Pakistan, on September 26, 1932. c India gained independence from the British on August 15, 1947. d India had a multiparty system, with six national parties and over 45 state-level parties. e The Indian National Congress, India’s oldest political party, played a leading role in India’s independence movement. f UPA tried to ensure economic growth, provide employment, and raise resources to implement poverty alleviation programs.
3
712-412
Aadhaar: India’s ‘Unique Identification’ System
The government looked to corporate India for leadership, instead of its administrative and political cadre, and on July 2, 2009, it invited Nilekani, CEO of Infosys, to be the chairman of UIDAI, with the rank of Cabinet Minister.g One of India’s leading businessmen was thus admitted to the inner sanctum of government, with enormous executive authority to accomplish nationally critical goals. This signaled a radical change in thinking from the “business as demon” approach of the license raj to recognition of the private sector’s talent.
Nandan Nilekani: Accidental Entrepreneur
Nilekani was the poster child of Indian meritocracy and clean governance. Born in a small town in southern India, Nilekani thought he was lucky that his father insisted he study in an English medium school, and even luckier to have graduated from the Indian Institute of Technology in Bombay. He started his career in 1978 with Patni Computer Systems. A disagreement with the Patni founders caused him to leave in 1981, with seven others, to start Infosys with $250. Infosys grew into a global leader in the “next generation” of IT and consulting, with revenues of $6.35 billion in 2011. Nilekani attributed this to luck as well: “I am but an accidental entrepreneur, who if he had not walked into the office of N.R. Narayana Murthy14 in late 1978 in search of a job, would probably have languished in a nine–to-fiver while living in a New Jersey suburb, taking the daily train to Manhattan.” 15
Nilekani co-founded the National Association of Software and Service Companies (NASSCOM) and the Bangalore chapter of The IndUS Entrepreneurs (TiE), and was awarded the Padma Bhushanh in 2006. In 2008, he wrote Imagining India: Ideas for the New Century, pointing out that India’s future depended on economic growth, reform, and innovation. His personal wealth was estimated in 2011 at
$1.3 billion; he was listed by Time magazine in 2009 as one of the world’s most influential people.
Nilekani and his co-founders built more than wealth. They were first-generation entrepreneurs with unrecognizable surnames. Nilekani said that Infosys created an “ethos of transparency and strong governance . . . [and] built a reputation for creating widespread shareholder wealth. People called . . . (Infosys) India’s ‘new economy’ company . . . they still call us that.” 16 Infosys built its own reputation; that of the Indian IT industry grew with it, and in 2011, India exported software to over 90 countries. In 1991, the Indian software industry garnered a mere $150 million, while revenues for
2013, despite a worldwide recession, were projected to reach about $100 billion.17
The success of India’s software industry fueled automation in banking, stock markets, land records, railway reservation systems, and electoral rolls, bringing transparency and reducing graft.
Nilekani noted, “For Indian consumers . . . IT has emerged as a tool to sidestep weak systems. People are realizing the ‘I’ in IT. Attitudes towards technology have undergone a sea change, from the hostility of 1980s and indifference of 1990s to an overwhelming demand for electronification today.” 18
Aadhaar: Foundation for Identity
In a world of global flows of wealth, power and images, the search for identity—collective or individual, ascribed or constructed—becomes the fundamental source of social meaning.
— Nandan Nilekani19 g Cabinet Ministers and the Prime Minister form the Cabinet of India, which is the highest decision-making authority in the
government of India. The rank empowered Nilekani with the highest executive authority. h India’s third-highest civilian award recognizes distinguished service of a high order to the nation, similar to Commander of
the Legion d’Honneur in France (see http://jssgallery.org/Essay/The_Legion_dHonneur/The_Legion_d'Honneur.htm).
4
Aadhaar: India’s ‘Unique Identification’ System
712-412
Approaches to identity management differed from country to country. Sharma explained, “The objective in Bangladesh was to clean up electoral rolls, in Pakistan to manage security, in the
European Union to control immigration, in the U.S to manage the flow of citizens or foreigners traveling across the border. UIDAI was the only program where the basic motivator was creation of identity for delivery of social services.” (See Exhibit 3.)
Errors of Exclusion: Need for Inclusive Growth
Sharma continued, “Many Indians don’t have a school leaving certificate or birth certificate.
People are not able to get an LPG [liquefied petroleum gas, used for cooking] connection . . . because they don’t have the documents to prove who they are.” i Recognizing that registration at birth was the first legal step to provide a child an identity, India made registration of births and deaths mandatory in 1969 and became a signatory to the United Nations Convention on the Rights of the Child in 1989.j
Yet, even in 2001, only 55% of births and 46% of deaths in India were registered. 20 One researcher said most children in rural schools had birth dates in May, June, or July, based on school enrollment dates rather than substantiated facts.21 Children without birth certificates faced hurdles in receiving healthcare, nutritional supplements, and school admissions, and had little protection against exploitation, premature enlistment in armed forces, or, if accused of a crime, prosecution as an adult.
Sharma revealed his dismay at the extent and impact of exclusion:
I come from a small village, Hamirpura, in Firozabad district of Uttar Pradesh (UP). I asked, as an experiment, who has an identifying document? Not more than 25% had any document
. . . My widowed sister-in-law, who was entitled to Rs 20,000 (US$400) from the Chief
Minister’s fund because her husband had died, could not cash the check. She had no bank account since she lacked the documents required by the bank’s Know Your Customer (KYC) norms.k I am convinced of the transformational nature of this project—not just for enabling service delivery, but in terms of its inclusion potential.
Government records indicated that 50 million Indians had passports, 95.8 million had the
Permanent Account Number (PAN) cards required for taxpayers, and 223.2 million held ration cards.22 Nilekani stated, “We will have 800 million people in our workforce by 2020. To ensure they are a source of valuable human capital, we must connect them to health, education and employment services. Aadhaar will enable us to implement our vision of ‘development with a human face.’”23
Errors of Inclusion: Multiples and Fakes
Most of the Indian social security services—pensions, scholarships, mid-day meals, food rations— distributed through the PDS were individual or family-driven. (See Exhibit 4.) Ration cards detailed how much subsidized rice, wheat, sugar, and kerosene a family was entitled to, depending on number of members. Sharma explained:
Leakage takes place where non-existent persons create documents, or one person creates more than one document. Suppose a family is entitled to 5kg of rice on one ration card—if I i LPG was subsidized, and hence allotments were restricted at a household level. j Article 7 of the UN Convention states, “The child shall be registered immediately after birth, and shall have the right . . . to a
name, the right to acquire a nationality, and as far as possible, the right to know and be cared for by his or her parents.” k As part of the Anti-Money Laundering Act, banks and cell phone providers had to follow a KYC process. They demanded a
photo and separate documents to prove identity and address, listing a choice of up to five documents for identity, and 10 for address. Ironically, a birth certificate was not acceptable, because it had no photo. In this way, the bank or mobile company verified an applicant’s identity. People without this proof were denied access to these services.
5
712-412
Aadhaar: India’s ‘Unique Identification’ System
create two ration cards, I get twice the 5kg. Or if I get two election ID cards, I can vote twice.
Or if someone wishes to avoid paying tax, they can open a benami account [Hindi for “bank account in a fake name”] using fraudulent documents . . . So essentially, incentives exist to create multiple or fake identities.
The inability to verify identity led to fraud. For example, examinees impersonated ill-prepared students for a fee, imposters drew pensions for dead individuals, and bigamists entered into multiple marriages with impunity. UIDAI thus needed to create an identity system that would provide identities to all, while ensuring uniqueness to eliminate duplicates and multiples. This entailed several challenges.
Challenge 1: Building the Organization
The UIDAI team decided that Aadhaar would be a unique portable identity, and would only guarantee identity, not rights, benefits, or entitlements. Sharma emphasized that “we want to limit ourselves to perform the function assigned to us; the project strategy is not to allow function creep.”
Nilekani was the first employee of UIDAI. This meant that he could create the organization he had envisioned. Sharma mused, “When you get a new task in the government you normally parachute into an existing organisation. It’s very unusual to set up office.” Nilekani elucidated, “We are a government organization—our budgeting, expenditure, reporting, accountability to Parliament . . . we have to comply with the Right to Information (RTI) Act and account for every penny of taxpayers’ money . . . But within that framework, we have tried to create an organization that has a sense of urgency, quality, and let’s-get-the-job-done kind of thing.”
One of Nilekani’s requests was to select at least a few team members. This was agreed to by the
Prime Minister’s Council of UIDAI, set up on July 30, 2009, to advise UIDAI on program, methodology, and implementation and to ensure coordination. Nilekani and Sharma divided their roles, with Nilekani focused on managing the environment and Sharma on organization building.
This ranged from obtaining office space, getting budgets sanctioned, and hiring people to building technology infrastructure. Since UID’s technical underpinnings would take a year or so to develop,
Nilekani used that time to travel: “I visited 25–26 state governments. I went to Raipur, Lucknow,
Patna . . . I met chief ministers and chief secretaries, made presentations, got everybody on board, met banks, insurance companies and oil companies. We had talks at industry associations. I spread the message.” His dream was to get every Indian a bank account and an e-mail ID.
Nilekani and Sharma conducted extensive interviews and background checks and collected references. UIDAI had people from organizations such as the Indian Audit and Accounts Service,
IAS, Railways, Defence, and Telecom (see Exhibit 5). The deputy director general of logistics at
UIDAI said, “In terms of perks, privileges and pay, this place has nothing more to offer than any government job. In many respects, it is worse off. Yet everyone is here because they want to [be].”
Recognizing that the project required high technical skills, UIDAI created a hiring model to attract talent from the private sector—people on sabbatical from Intel, McKinsey, and General Electric; interns from across the world; consultants; and volunteers. Government pay scales were lower than those in the private sector, so it was the scale and objective of the project that was the draw. One volunteer, a Harvard Business School MBA student, shared, “Large-scale transformations require policy changes . . . For someone coming in with a career in Silicon Valley and background in entrepreneurship, it is virtually impossible to be involved in a key policy initiative. This was a unique opportunity to be an insider in what is probably the most sweeping IT initiative in India.”24 Raj
6
Aadhaar: India’s ‘Unique Identification’ System
712-412
Mashruwala, a 58-year-old Silicon Valley investor, volunteered to relocate to Bangalore after 36 years in the U.S. By December 2010, around 30% of the team of 200 was drawn from unpaid volunteers
(23), those on sabbatical (14), and others (23) who had taken a steep pay cut to be part of the Project
Management Unit (PMU).25
Government officials headed all departments and made policy decisions, aided by experts from the private sector. Nilekani reflected, “This motley mix brought some tension, but also dynamism.”
Those from the government were used to hierarchy and paper trails. Those from the private sector were used to first names and e-mails, and didn’t understand that every penny had to be accounted for. Nilekani insisted there would be no silos; Sharma introduced weekly meetings and videoconferences. Nilekani was no stranger to scale. However, he confessed, “I was blown away by the size, scale, reach and impact of the government . . . The state government will say, ‘we will finish this in one year,’ and you say ‘they are talking about enrolling 100 million people in one year!’” One volunteer commented, “[Nilekani] assembled a star team from different disciplines . . . For people from the private sector; it was an exercise in understanding how government policy making works.”
Challenges remained. The GoI had constituted a Cabinet Committee on October 22, 2009, headed by the Prime Minister, which included ministers from every function that UIDAI would interact with.
Ahluwalia and Nilekani were special invitees. l Yet, forging such diverse working relationships was not easy. One of the deputy director generals explained, “How do you deal with an issue in which you are not the prime mover, but want to play a role without antagonizing other stakeholders?”
Nilekani observed, “We synthesized the best of both worlds . . . Government is a different world, but a lot of principles are similar. In IT consulting, when you call on a customer, you try to understand his business and provide solutions. When I call upon potential users of Aadhaar, I try to understand how can it make their activity better . . . That win-win approach to relationship building is what we were able to bring to the table . . . Rather than build a huge organization, we will be the small organization and partner everyone.”
Sharma said, “Every bureaucratic organization wants to spread.” The initial proposal was to build an organization with UID commissioners in each state and a plethora of supporting staff. This was rejected as being unwieldy. Instead, UIDAI decided to utilize existing infrastructure. “We are 200 people, only 50% of our sanctioned strength; our approach is to create an ecosystem rather than build the whole structure. We should centralize issuing of numbers and decentralize enrollment stations.”
Nilekani elaborated, “We have three ecosystems. One is the enrollment ecosystem that allows us to scale up to a million a day by creating enrollment agencies; second is the application ecosystem that allows us to build apps; third is the device ecosystem, where we have created application programming interface (API)-based approaches.”26
Challenge 2: Building the Technology
Aadhaar was the government’s largest and most technology-savvy initiative, and the GoI envisaged development of a technology backbone to support three aspects—enrollment, deduplication, and online authentication—while ensuring information security. (See Exhibit 6.) A key
l The Cabinet Committee included the Ministers of Finance, Agriculture, Consumer Affairs, Food and Public Distribution,
Home Affairs, External Affairs, Law and Justice, Communications and IT, Labor and Employment, Human Resource
Development, Rural Development and Panchayati Raj, Housing and Urban Poverty Alleviation, and Tourism.
7
712-412
Aadhaar: India’s ‘Unique Identification’ System
member of the team that built Intel’s Pentium processor was recruited to head of the Technology
Development Unit (TDU). The TDU attracted experts in hardware, software, devices, and legal issues.27
UIDAI decided to link the 12-digit randomly generated number—Aadhaar—to biometric data including a photograph, all 10 fingerprints, and iris scans of both eyes, and to demographic information including name, address, date of birth, and gender.28 Fingerprints eased authentication, but iris scans were also needed for infants and those whose fingerprints were worn away by physical labor. UIDAI set up a Biometrics Centre of Competence to build a key group of scientists who would implement biometric-compatible systems across departments. Biometric databases of the magnitude
UIDAI was considering (that could cover 1.2 billion people) were unprecedented. By comparison, the
Federal Bureau of Investigation (FBI) in the U.S. had a database of fingerprints and criminal histories for 66 million criminals, as well as 25 million civilian fingerprints.29 One of the deputy director generals shared, “When on a visit to New Zealand, I mentioned that UIDAI . . . had covered 4.5 million enrollments. Their jaws dropped—they said at 4.5 million, we had covered the entire population of New Zealand.”
Enrollment Equipment
The enrollment equipment, packed in a briefcase, comprised a laptop (with enrollment software, fingerprint reader, and iris scanner), webcam, laser printer, and monitor30 (see Exhibit 7). Meanwhile, the cost of an iris camera had decreased from $254131 to $782.32 An officer said, “Enrollment occurs in two languages—English and the regional language. As transliteration occurs simultaneously, the enrollee can check data accuracy.” After enrollment, the enrollee received an Enrollment ID (EID) that could be used as a tracker until the Aadhaar was received. Data from laptops was sent in encrypted form to data centers with high levels of security. For de-duplication, all biometric data was compared to existing IDs. To ensure privacy protection, responses to authentication queries could only be “yes” or “no.”33 Another commented: “Everyone thought biometric capture required special conditions. We proved this could be done with minimal conditions of light, in varied temperature conditions and environments—rural, urban, slums, and night shelters. [See Exhibit 8.] We took some good decisions—that the data would be stored on the computer, not online, and the process would be designed for people with very simple IT skills.”
Enrollment Process
To enroll, one had to produce proof of identity, address, and date of birth. For those who could not verify their date of birth, the date they chose was considered the final date. 34 Designated introducers could introduce those who did not have proof of identity or address; children below 15 had their number issued based on their parents, and had to re-register their data and renew the number after age 15. It was envisaged that Aadhaar would be inserted into birth certificates. (See
Exhibit 9.) On death, the record would be marked deceased, but not deleted.
UIDAI built partnerships with various “registrars” to enroll residents. Registrars are organizations that interfaced with large numbers of people—e.g., that managed the Public Distribution System
(PDS) or the MGNREGA scheme, or distributed LPG. By September 2011, UIDAI had 69 registrars, which included state governments, public sector units, the Life Insurance Corporation of India,
National Stock Depository Limited, public sector banks, India Post, and telecom companies. Registrars had to set up enrollment camps and were encouraged to partner with agencies to enroll residents.
8
Aadhaar: India’s ‘Unique Identification’ System
712-412
De-Duplication in the Database System
Before a fresh Aadhaar was issued, the biometrics were compared across all prior records, and only when there was no match were the details inserted into the database and a unique number generated.
One commentator wrote, “[UIDAI] will photograph 600 million Indians, scan 1.2 billion irises, collect six billion fingerprints and record 600 million addresses . . . No system in the world has handled anything on this scale . . . When the 600 millionth person is assigned a UID, the system will have to compare it against 599,999,999 photographs, 1,199,999,998 irises and 12,999,999,990 fingerprints . . .
When in full flow, it will add a million names to its database every day.”35
Online Authentication—Anytime, Anywhere, Any Way
A member of the PMU explained, “When a resident visits a ration shop, the information containing his Aadhaar and biometric-fingerprint will come to the Central Identity Data Repository
(CIDR). The CIDR will say, yes this person is who he claims to be. That is authentication. Deduplication is a 1:n match, is process-intensive, but happens only once for every resident.
Authentication is a 1:1 match, access to every service requires authentication, so it is transactionintensive.”
One deputy director general said that the real test lay in accurate, timely authentication: “Our goal is to send back a Yes/No response within eight seconds.” UIDAI’s centralized infrastructure enabled
“anytime, anywhere, any way” authentication and could be done offline and online. Online authentication allowed residents to verify their identity remotely and gave rural residents the same flexibility that urban people had in verifying their identity. (See Exhibit 10.)
Another deputy director general elucidated, “Authentication processes worldwide are based on three items—what you are, what you have, and what you know. ‘What you are’ is your photo, your biometrics. ‘What you have’ is your ATM card or user ID. ‘What you know’ is your secret PIN or password. Worldwide, most authentications are based on ‘what you have’ and ‘what you know.’ No system uses ‘what you are’ biometric authentication on the scale and scope we are planning.”
Existing authentication systems operated by Visa and MasterCard, while massive, were not strictly comparable, since they were card-based. MasterCard processed 23 billion transactions a year, while Visa processed 47 billion a year.36 That meant that Visa processed 130 million transactions a day, or 20,000 transactions per second. At even the interim goal of 600 million residents, and assuming only one-third performed a transaction every day, the UIDAI authentication system would be handling more transactions per day than the world’s most sophisticated payment system.
Challenge 3: Changing Behavior, Dealing with Resistance
UIDAI introduced radically new ways of doing things, requiring many to change their mindsets.
Mandatory versus Voluntary
Sharma said, “If Aadhaar confers no entitlement, why should anyone sign up for this? If I make it compulsory, I need to create a legal framework, and this will become a barrier to accessing benefits
. . . At first there will be no one with Aadhaar; so if I mandate that PDS will deliver subsidized food and fuel only to those with Aadhaar, I shut out everybody and defeat my purpose.” Nilekani mused,
“Demand-led and voluntary is the only sustainable way. Making it mandatory does not mean it will happen. Partner agencies can make their own use of Aadhaar mandatory.”
9
712-412
Aadhaar: India’s ‘Unique Identification’ System
The first Aadhaar was issued on September 30, 2011, to Ranjana Sonawane, a 30-year-old housewife from a village in Maharashtra.37 Enrollment trends picked up quickly. (See Exhibit 11.) But the demand for Aadhaar took the team by surprise. Long lines formed at enrollment stations, and reports complained about the wait time, slow process to capture data, and limited capacity of each center. People were worried that the stock of Aadhaar would run out. More poignantly, an emaciated
30-year-old said, “I came to Mumbai from a UP village 15 years ago. I work as a watchman and sleep in the building parking lot. I send money home, but since I don’t have a proper home, I can’t open a bank account. I have to ask a friend to take the money back. Sometimes it doesn’t reach my family . . .
With Aadhaar, I can open a bank account, save and perhaps rent a room to stay in.” 38
Sharma and Nilekani used these hiccups as opportunities. They took steps to improve the process, signed up more registrars, entered into agreements with National Skills Development Corporation
(NSDC) to train more operators, published advertisements making it clear the application was free, and introduced an appointment system to remove wait times. (See Exhibit 12.) Sometimes, however, delays persisted.
Service Providers and Their Concerns
Sharma clarified that enrollment for Aadhaar was a one-off activity and short-term goal. The real challenge lay in getting multiple service providers to use Aadhaar; the latter would provide long-term sustainability. “So far, all IDs that Indians have used have been pre-linked to service delivery—ration card for subsidized food, the MGNREGA job card number, which is linked to jobs. [See Exhibit 4.]
It’s difficult for residents and service providers to understand that Aadhaar is not linked to service, but to ‘what you are.’”
The concept of being in control of one’s own data was so ingrained in most organizations that they could not fathom alternatives. A member of the PMU said that signing up service providers was not about explaining the concept; it required process re-engineering to enable their databases to accept
Aadhaar. That was another hurdle, since “anytime you ask people to modify their application, they say, ‘that’s a big task,’ whereas it’s not that big a deal. So we are creating implementation models to convince others to replicate the same.”
One deputy director general said UIDAI insisted that Aadhaar be declared sufficient for KYC norms. “However, if the state government says ‘help us redesign PDS around Aadhaar,’ or ‘help design disbursement of pensions,’ we work with them. Some say we are promoting Aadhaar as a panacea. That’s not true. Since it’s demand-led, we have to reach out. Then activists say, ‘why are you marketing?’ I think the government lives off criticism, just like the private sector lives off profit.”
Migrants, Homeless, and Illegals
Aadhaar was likely to be the first form of identification that the poor, underprivileged, migrants, street dwellers, and homeless would access. UIDAI did not want any barriers in enrolling them.
UIDAI borrowed the concept of an introducer from banks’ account-opening procedures and created
Know Your Resident (KYR) standards with an introducer system. Authorized individuals
(“introducers”) who had Aadhaar could introduce residents who did not have any documents.39
Introducers were wary of doing this, which led to sensitization programs clarifying that introducers were vouching only for identity and address, and not for character, antecedents, or criminal records.
A deputy director general shared her experience of a pilot in Delhi to enroll the homeless: “I am often unable to deliver Aadhaar to the homeless. Imagine how many haven’t even got enrolled, because they haven’t been surveyed. Every time I drive by Nizamuddin basti [Hindi for ‘slum’], I feel
10
Aadhaar: India’s ‘Unique Identification’ System
712-412
like asking—have you got enrolled or surveyed? . . . I can’t do a media awareness campaign for the homeless (because they don’t watch TV or read the newspaper) . . .Reaching them occurs only through non-profit organizations, but that’s not always successful.”
While UIDAI maintained that Aadhaar did not confer citizenship, there were fears about illegal residents unfairly claiming entitlements.,40 Others argued that counting illegal residents was the first step in doing something about them. One deputy director general declared, “It is not our mandate to check citizenship at the point of enrollment. If there is illegal immigration, there is the Indian army, paramilitary forces and police . . . What can we do about it? Our job is the ID management program.”
Collecting Information: Privacy Concerns
UIDAI consciously limited the amount of information that it collected. Sharma revealed:
There were two options. One was collect as much as you can. Details on health, spouse, kids, income, tax, loan, pension payments, bank account and electoral rolls were tempting to collect . . . Another option was, why should I collect anything more than what’s required to prove identity . . . So, basically, there was a tussle between loading the database so it could support many applications, and limiting ourselves to identity management. Each route has its advantages and disadvantages.
UIDAI formed a committee of stakeholders and decided to collect minimal demographic data— name, age, gender, and address. Sharma explained, “Else people will say, why are you collecting extra information? You will misuse it.” Criticism continued, with blogs describing hypothetical scenarios where identity theft via Aadhaar left people bankrupt, jobless, denied medical insurance and school admission, and accused of theft. 41 Nilekani defended Aadhaar, saying that data was kept offline, and access restricted. “This is the world’s first online ID system. From no ID to online ID—we are jumping two steps.” He acknowledged the need for a data security law. In April 2011, the
“Information Technology (Reasonable Security Practices and Procedures and Sensitive Personal Data or Information) Rules, 2011,” or “Privacy Rules,” were issued to implement India’s 2008 IT Security
Act amendment. These rules imposed limitations on businesses’ use of personal data.42 However, concerns remained amid reports of the government tapping e-mails and phones.43
The UIDAI team recognized that if it had collected more data, applications would have been easier to create. However, team members believed that the cost of updating and maintaining the data would have been enormous, and would have distracted them from their core purpose of proving identity. Sharma argued, “My job is to say X is X. So we decided that UIDAI will not own and maintain this data, and that the entities who collected the biometrics and data for us could keep the additional information if they needed it.”
Execution and Implementation
Enrollment Ecosystem: Reach, Accuracy, and Delivery
An IAS officer now at UIDAI, commented on the partnering philosophy: “We had to lay down standards so that different partners—state governments, banks, private sector players—acted in consonance.” UIDAI was structured into eight regions, each region covering a population of around
160 million. The Bangalore office, for example, represented one such region covering Karnataka,
Tamil Nadu, Kerala, Pondicherry, and Lakshadweep. The regions were responsible for front-end
11
712-412
Aadhaar: India’s ‘Unique Identification’ System
enrollment, signing a Memorandum of Understanding (MoU) with the state government and the registrars, and appointing enrollment agencies via a bidding process. Registrars had to ensure that correct data was collected, as well as aggregate enrollments from sub-registrars and enrolling agencies and forward them to CIDR. Each registrar had to adopt UIDAI standards in the technology used for biometrics, as well as collect and verify resident information and submit to audits. 44
The enrollment ecosystem was like a three-legged stool requiring registrars, enrollment agencies, and operators who enrolled residents by collecting data. UIDAI signed MoUs with 69 registrars, impaneled over 200 enrollment agencies and 15 training agencies, trained 70 master trainers, and created training material and a certification process for operators. Registrars could appoint any of the impaneled enrollment agencies. Hiring operators was the enrollment agency’s responsibility.
An assistant director general estimated that to meet operator system goals would require 50,000 operators to be trained and certified. By June 2011, 15,000 operators had been certified, but only 5,000 were on the job. Productivity varied; some operators could enroll 60–70 residents a day, others only
10. Analysis revealed other issues. One officer commented that the environment where operators worked was not always favorable, skill levels varied, and “Gross errors were noticed at CIDR. This is a complex project . . . we decided to make only certification mandatory, not training.”
“If productivity of operators goes up, everyone wins,” said an assistant director general. “The model didn’t take off too well, so we revised it. We developed operator-level reports, which we shared with the enrollment agency. We insisted that data collected is reviewed by the enrollment agency before it is sent to CIDR. We introduced coaching . . . we are working to see if a training ecosystem can be created.” Enrollment peaked in January 2012 at over 24 million enrollments a month,45 stretching the capability of the postal system to print and deliver the letters.
Enrollment slipped to less than six million in the month of July 2012.46 Part of the issue was a clash between the Home Ministry’s NPR (National Population Register) and UIDAI on scope and approach. The NPR objective was to issue national identity cards to Indian citizens, thus excluding illegal immigrants. The UIDAI mandate was to issue Aadhaar numbers to all residents of India; the number conferred neither citizenship nor entitlements. Government institutions would be able to use
Aadhaar as a foundation to accurately identify and include poor in the benefits of development. One commentator wrote, “One [NPR] is rooted in a mindset of exclusion and security, the other [Aadhaar] is inclusive and participative . . . these divergences . . . explain differences in issues like data points, methodologies and secure storage of data . . . [but do] not explain the constant undercurrent of hostility that exists between these two programs. Nor [do these divergences] explain why the popular discourse on these two programs rests exclusively on ‘duplication’ and additional cost to the exchequer.”47 A compromise was finally reached, with UIDAI being given a budget of $2.7 billion48 to enroll a further 400 million residents by June 2014. UIDAI and NPR agreed to take the responsibility for collection of biometric data in different states, and accept each other’s data.
The New Yorker magazine stated that Murthy (Nilekani’s first boss and co-founder at Infosys) believed “the political obstacles were always larger than the computing ones.” Murthy noted, “The challenge is in making sure that literally hundreds of thousands of officers fall in line, they rally to his
[Nilekani’s] call and march to his tune. I would not hold him responsible if this project did not take off, if it did not scale up as well as he wants. Most things in India muddle along.”49
Application Ecosystem: Usage and Impact
Building on the acceptance by banking and telecom authorities that a resident with Aadhaar fulfilled KYC norms, pilot projects to test authentication services using the data centers in greater
12
Aadhaar: India’s ‘Unique Identification’ System
712-412
Noida (near Delhi) and Bangalore began in late September 2011. By August 2012, regulators agreed to accept Aadhaar as proof of both identity and address, enabling investors to, for example, buy mutual funds.50 Nilekani expounded:
We have built an enrollment ecosystem that gives us scale and speed. Long-term sustainability will come from the application ecosystem . . . We want to create a virtuous cycle, where applications are created because many are enrolled, and people enroll because of the applications. While initially we see telecom and financial inclusion as two big benefit drivers, that’s just a start. We are working with state governments and ministries to encourage them to develop government applications. We have a budget to feed these application developments, and we are giving a governance award to district collectorsm who come up with interesting apps. UIDAI prepared reports on Aadhaar-enabled service delivery and applications for government schemes that subsidized education, public health, food, fuel, and rural work.51 The prime minister, finance minister, and the president of India endorsed the role of Aadhaar in reducing leakages and improving delivery of services to the poor while presenting the budget to Parliament in conferences and speeches.52
India regulated the price of domestic cooking fuel to guard against international price fluctuations, and subsidized it to give access to the poor. The customer base of ~133 million households (each household was allowed one connection) was served through a network of over
11,000 LPG distributors, which delivered around three million LPG cylinders every day. The total cost of the subsidy for LPG in 2010 was estimated to be $4.8 billion, a heavy burden for the national budget—yet targeted beneficiaries often failed to receive their benefits.53 The arbitrage between the market price and the subsidy, the versatility of the fuel, and the weak verification systems for delivery led to rampant diversion through fake and multiple identities. 54 The Minister for Petroleum and Natural Gas said, “To avoid leakages in the system, the government is working out modalities for cash transfer of subsidies using Aadhaar platform.”55 In January 2012, a pilot project for distribution of domestic LPG cylinders using Aadhaar was rolled out through three distributors in
Mysore, because Aadhaar coverage there was more than 90%.
The experiment to integrate Aadhaar with MGNREGA in three districts of the state of Jharkhand in January 2012 was more controversial. The objective was to make access to banking easier for rural workers, bypassing post offices, where identity fraud was endemic. Banking correspondents (BCs), who were bank-appointed agents, paid MGNREGA workers their dues. BCs identified workers through Aadhaar-based biometric authentication using a handheld device called a micro-ATM.
(See Exhibit 13.) One critic who visited the project site pointed out that fingerprint recognition,
Internet dependence, and general process failures (including missing Aadhaar numbers, lack of bank passbooks, or no bank account) made the experiment less than successful, though the critic did admit,
“the system seemed to work . . . most workers appreciated being able to collect their wages closer to their homes, without the hassles of queuing in overcrowded banks or of depending on corrupt middlemen to extract their wages from the post office. They did not fully understand the new technology, but nor were they afraid or suspicious of it.”56 The article prompted more than 35 comments ranging from agreement to objections, and a spirited rejoinder from the Minister for Rural
Development, who claimed in his op-ed, that “Aadhaar and MGNREGA are made for each other. . . .
Process re-engineering in government is never easy; there is enormous resistance both from within and outside the system. Aadhaar has the potential to be superior to other biometric solutions.”57 m District collectors, as administrative heads of a district, were the most powerful government officials of that region.
13
712-412
Aadhaar: India’s ‘Unique Identification’ System
Implementation challenges were recognized, but the approach that was adopted was to
“experiment, learn and innovate” and to extend the use of Aadhaar in MGNREGA to all districts where enrollment was nearing 90% to 100%. It was expected that over time, MGNREGA cards would carry the Aadhaar number, details of BPL status, and government insurance schemes, enabling convergence of social schemes and Aadhaar-based verification of identity.
Fledgling applications by entrepreneurs around Aadhaar also surfaced. Twenty-three-year-old
Shardul Lavekar, co-founder of EasyMontra (short for “easy money transaction”), described how his company was facilitating doorstep banking in a remote village in Maharashtra: “Our four-member team of engineers has developed a solar powered handheld device that can be used as a microATM.” The device was named M-Artha (Artha was Sanskrit for “wealth”; the name meant “mobile wealth”)58 and consisted of a touchscreen, fingerprint sensor, RFID reader, general packet radio service (GPRS) and global positioning system (GPS) facility, and a thermal printer. It could perform eight types of transactions, including cash deposits and withdrawals, loan repayments, utility payments, and money transfers.59
Lavekar explained, “We had planned to use Aadhaar for authentication of identity, but found that the village had very few enrolled. Hence, we have had to build our own biometric database; for cost reasons we collect only fingerprints [iris scanners were expensive], but often, manual labor erodes the quality of these.” He explained that EasyMontra would have preferred to pilot its device in a village with good Aadhaar coverage, because it would have made things easier, but the company was limited by the bank’s mandate. Lavekar continued, “One advantage of building our own biometric database is that authentication is really quick—a second or two, probably because the database is quite small.” The delay in the rollout of the Aadhaar API, and uncertainty around enrollment, spurred EasyMontra to look for a mobile phone PIN-based solution. EasyMontra was one of the winners of the business plan competition on Aadhaar applications, which drew over 60 applicants, held at a leading Indian business school.60
Conclusion
The journalist talking to Nilekani persisted, “How will you know you have succeeded?” Nilekani responded, “Five years from the day I [began], you would be able to say I succeeded if people got numbers, and you’d know I screwed up if people didn’t get numbers. So it was a zero or a one.”61
The journalist paused, then said, “UIDAI needs to . . . collect 12 billion fingerprints, using the 17 languages on a Rs 10 note. And you plan to do this while creating a better-than-ordinary experience for every resident? It’s an epic project . . . not unlike the challenge of governing the world’s largest democracy. And it’s not easy—Indians are losing patience.”
“What UIDAI is creating is a road,” responded Nilekani. “A road that connects every individual to the state. How each one of us uses that road, how far we travel along it, is up to each one of us. We have to look forward.” He added, “The best way to deal with (opposition) is to show its usefulness . . . to give unmatched benefits . . . that is what will win the debate. “62
The journalist, all of 25, stood up and said, “Yes, there are concerns; such projects can so easily be derailed. The question remains: How can UIDAI de-risk the project so that Aadhaar—Aam Aadmi ka
Adhikaar [right of the common man] is made available to every resident of India by 2020?”
