NEW DELHI — The Narendra Modi government is in the final stages of creating an all-encompassing, auto-updating, searchable database to track every aspect of the lives of each of India’s over 1.2 billion residents, previously undisclosed government documents reviewed by HuffPost India establish.
If the plans of Modi’s bureaucrats and advisors are realised, this system will automatically track when a citizen moves between cities, changes jobs, buys new property, when a member of a family is born, dies or gets married and moves to their spouse’s home. The interoperability of modern database systems means there is no technical limit to the extent of data that can be collected and indexed by this master database of databases. In a meeting on October 4, 2019, for instance, a special secretary of the NITI Aayog even proposed geo-tagging every single home and integrating it with Bhuvan, a web-based geo-spatial portal developed by the Indian Space Research Organisation (ISRO).
Five years in the making, the proposed National Social Registry has thus far been described by the Indian press as a routine exercise to update the 2011 Socio-Economic Caste Census (SECC) to prevent the misuse of pro-poor government schemes by ensuring that benefits and entitlements reach the right people. The fact that the Ministry of Rural Development is responsible for the SECC has helped further this impression that the SECC update is an innocuous bureaucratic task.
Now, documents obtained through the Right To Information Act by Srinivas Kodali, data and internet governance researcher, and this correspondent suggest quite the opposite: Under the guise of creating a SECC that automatically updates itself in real time, the National Social Registry (or the Social Registry Information System or SECC Social registry as it is also known) will either be a single, searchable Aadhaar-seeded database or “multiple harmonised and integrated databases” that use Aadhaar numbers to integrate religion, caste, income, property, education, marital status, employment, disability and family-tree data of every single citizen.
The National Social Registry, the documents make clear, will not be restricted to sucking up data on below-poverty-line families who rely on state support, but every single Indian citizen.
Unlike the Indian population census — governed by the Indian Census Act of 1948 that legally mandates the confidentiality of those enumerated — the SECC has no such safeguards.
Yet this registry is no pipe-dream.
File notings, meeting minutes and interdepartmental correspondence reviewed by HuffPost India reveal that the government has taken concrete steps towards building this database:
An expert committee has been set up to implement the social registry by next year, i.e 2021. The committee is in the final stages of planning a pilot project to test the best way to get it done.
This expert committee has proposed amendments to the Aadhaar Act to allow the government to capture this information without running afoul of the 2018 Supreme Court judgment that restricted the use of Aadhaar and reiterated individual privacy as a fundamental right.
The Unique Identification Authority of India (UIDAI) has decided to amend the Aadhaar rules, the minutes of the October 4 meeting show. If implemented, these changes will render the 2018 judgment meaningless by removing the few privacy safeguards that are part of the original Aadhaar law.
UIDAI has suggested a “Data Exchange Framework” to ensure that the hundreds of government-administered databases scattered across several ministries and departments at the state and central level can easily exchange data. HuffPost India could not establish if this proposal has been accepted.
The World Bank has also “assured cooperation”, according to a file noting dated June 17, 2019, and has agreed to an initial $2 million grant under the bank’s Non-Lending Technical Assistance programme.
These revelations are particularly significant at a time when India’s Home Minister Amit Shah of the Bharatiya Janata Party (BJP) has openly spoken of creating a National Register of Citizens to weed out so-called foreign infiltrators that Shah claims are hollowing out India like “termites”. Shah has never provided any evidence to support his claims.
If the registry is implemented in its current form, privacy experts warn, the government can use opaque algorithms to sift through reams of data and arbitrarily designate individuals as citizens or non-citizens. HuffPost India has previously reported on how the use of such algorithms may have robbed millions of citizens of their right to vote in the Telangana and Rajasthan state elections in 2018.
The SECC’s decade-long mutation, from a foundational census to ostensibly efficiently deliver benefits to the poor into an mass-surveillance system, reveals how the Indian government is rapidly building surveillance infrastructure under the guise of poverty alleviation and how organisations like the World Bank are happy to offer advice to developing countries building intrusive systems that may not pass muster in places like Europe.
“Such an unrestrained mass surveillance system could threaten liberty like never before,” said Chinmayi Arun, Fellow of the Information Society Project at Yale Law School, who taught law in India between 2010 and 2018.
“India’s safeguards for state surveillance have always been weak. But this near-complete Orwellian surveillance would overturn the balance of power between citizens and the state,” Arun said. “It may be safe to say that if the state manages successfully to watch us so closely, India’s democracy will gradually become unrecognisable.”
A ‘dynamic’ registry
In 2011, Prime Minister Manmohan Singh’s United Progressive Alliance government embarked on India’s first caste-based census since 1931. The 2011 SECC, as its name makes clear, was a census, not a survey: this meant the government sought to collect data on the caste, incomes and set social parameters of every single Indian citizen.
The project was coordinated by the Ministry of Rural Development but conducted by three different government agencies: The ministry of rural development handled rural India, the urban census was carried out by the Ministry of Housing and Urban Poverty Alleviation, and the politically-sensitive caste census was administered by the Ministry of Home Affairs: Registrar General of India (RGI) and Census Commissioner of India.
On July 3, 2015, the Bharatiya Janata Party (BJP)-led government published the socio-economic data captured by the SECC, but withheld the politically-sensitive caste data.
The SECC heralded a quantum shift in the way the Indian state understood the question of poverty and entitlements. For decades, Indian policymakers had defined the poor as families whose annual incomes placed them below a threshold defined as the “poverty line”.
The SECC sought a more granular picture of entitlements to capture the increasing financial complexity of the lives of Indian citizens, and the increasingly focused nature of government support schemes. State and central governments were now rolling out schemes that looked at indicators other than annual family income — like school scholarships earmarked for girl students, loans to start small businesses, etc.
The SECC data, Chaudhary Birender Singh, the Union Minister for Rural Development at the time, said, “addresses the multi-dimensionality of poverty and provides a unique opportunity for a convergent, evidence based planning with a Gram Panchayat as unit.”
On 13 October, 2015, the Ministry of Rural Development proposed a social registry system to “ensure greater benefits from SECC data” to the Parliament Standing Committee on rural development, according to a November 2015 file noting reviewed by HuffPost India.
On November 27, 2015, the then-Economic Advisor of the Ministry of Rural Development, Manoranjan Kumar, composed a detailed note that would serve as the basis for an ambitious project to turn the SECC data into a continually updating registry.
“To be effective social registry SECC would need continuous updating to become dynamic social registry,” Kumar wrote in his note.
The system should update itself automatically, Kumar’s junior Dhruv Kumar Singh suggested. “The proposed system is subject to auto-updation in future since the profile of beneficiaries will change once they receive any support.”
That way, the government would know when a family on the brink of impoverishment suddenly became eligible for assistance, or conversely, when state benefits offered to a poor family might have improved their financial security to the point they were no longer eligible for state support.
Further, families were constantly moving from rural to urban areas, and back again — implying that the distinction between urban and rural families (eligible for corresponding urban and rural welfare schemes) was becoming harder to maintain.
Kumar’s solution was to come up with a centralised database, or registry, that relied on Aadhaar-enabled transactions to capture vast swathes of information about every Indian.
“The MoRD needs to opt for the largest set of database (all the households in the country) if the country has to deal with poverty in a non-asymmetric manner,” Kumar wrote.
Where government departments had previously maintained lists of scheme beneficiaries, the Social Registry would capture what Kumar called “potential beneficiaries” — and since everyone in India could potentially, at some point, avail of some government scheme, the system would capture everyone.
Kumar’s idea was in line with the government’s increasing focus on hi-tech solutions, led by Aadhaar, to deliver welfare schemes. Experts believe these solutions have not worked the way the government claimed they would.
“The past experience of when the government uses data and hi-tech tools to target the right beneficiaries has not been good or efficient,” said Nikhil Dey of Mazdoor Kisan Shakti Sangathan, an organisation that campaigns for transparency in implementation of welfare schemes. “The power structures provide wrong or fake information to ensure their own inclusion, at the cost of the poor and marginalised.”
“What the government has done with the tools like Aadhaar is that they have excessively collected information and have then used privacy as an excuse to keep it away from the community,” Dey said. “The potential of misuse of techno-managerial tools is huge.”
Kumar’s note triggered a five-year long process that continues to this day. Over the years, various departments, ministries, think-tanks and agencies like the NITI Aayog, the Unique Identification Authority of India (UIDAI) and the World Bank have weighed in with suggestions.
In theory, a dynamic database that helpfully points out everyone who needs government assistance sounds like a great idea. But the only way to do so in practice is by continuously monitoring the economic and social lives of every citizen — something that experts with the best intentions did not realise at the time. Over three years, every additional suggestion has only made the Social Registry more and more intrusive.
In January 2016, for instance, the Ministry of Rural Development constituted an expert group under the former finance secretary Sumit Bose to define criteria for identifying beneficiaries of various welfare schemes using the SECC. The committee submitted its report endorsing the idea of a dynamic database in November 2016.
Yet, some members of the committee said they would never have endorsed a database as intrusive as the Social Registry appears today.
“Such profiling of the entire population, through seeding of Aadhaar, was never our recommendation,” said Himanshu, associate professor of economics at Jawaharlal Nehru University (JNU), Delhi, who was part of the Sumit Bose committee that recommended creation of the social registry in 2016.
“Our suggestion was to create one common register of all eligible families for all the subsidies and welfare schemes using the SECC, which can be constantly updated, so that there is convergence of the schemes,” Himanshu told HuffPost India in an interview. “But the government never got back to us.”
In March 2016, the government had also roped in the NITI Aayog to offer their suggestions.
The NITI Aayog said “the concept of family tree” must be built into the information system for “added advantage”, according to a note dated May 13, 2016, penned by Senior Statistical Officer S.C. Jha.
For the Social Registry “to be relevant all times and its utility for various government programs” the NITI Aayog said it must be linked to birth, death and marriage Registers, to “account for migratory changes”. The ministry agreed to “incorporate” the suggestions, shows a note dated May 20, 2016 by Dhruv Kumar Singh, director in the ministry.
Meanwhile, negotiations and conversations between the government and the World Bank’s India office continued apace. But by March 2017, some officials in the Ministry of Rural Development appeared unimpressed by the Bank’s suggestions.
“All options offered by the WB are feeble cases, far weaker than many states in India have to offer,” Manoranjan Kumar, the economic advisor to the Rural Development Ministry at that time, wrote on March 15, 2017. He suggested the ministry should look at the Social Security System of the USA as a model, “which not only helps track the economic status of individuals but also traces individual’s interaction with any government program.”
In June 2017, the ministry constituted an Inter-Ministerial Expert Committee for “examining the feasibility of updating the SECC 2011 data” and “suggesting institutional framework for managing the Social Registry”.
The committee had members from the UIDAI, the World Bank, National Informatics Centre, the Ministry of Statistics and Programme Implementation, Ministry of Housing and Urban Affairs, Centre for Digital Financial Inclusion and the Direct Benefit Transfer Mission.
The committee, the file notings suggest, has met four times between June 2017 and October 2019.
The Ministry of Rural Development, the UIDAI and the Registrar of General of India under the union home ministry did not respond to HuffPost India’s queries despite repeated reminders.
The World Bank confirmed its “engagement” with the Rural Development Ministry on the Social Registry. The Bank’s support “did not warrant any lending” and involved “limited exchange of knowledge”, a spokesperson said in an email to HuffPost India.
“Violation of privacy and misuse/surveillance are of utmost concern for the central and state governments and the Bank,” the Bank said. “As part of the Technical Assistance, the World Bank shared multiple examples and approaches of different countries highlighting the importance of privacy, protection and sharing of data.”
The bank, however, refused to share its submissions to the government, saying they were in “deliberative stages.”
On March 5, 2018, the Prime Minister Modi signed off on a proposal to update the SECC. The specifics and safeguards of the SECC were yet to be ironed out, but one thing was clear — it was now a question of when the registry would be built, rather than if it would be built at all.
The Aadhaar effect
Were it not for Aadhaar, the Social Registry would never have gotten off the ground. The presence of Aadhaar as a “single identifier” has made it easy to merge several databases into a single registry.
For instance, imagine two separate lists — of PAN numbers and cell-phone numbers — each seeded with their respective unique Aadhaar numbers. It is now easy to create a unified database of PAN numbers and phone numbers using Aadhaar as the common identifier.
But there was a hitch.
Privacy experts had filed a series of petitions in the Indian Supreme Court warning of precisely such a scenario, where the mandatory seeding of Aadhaar in everything from airport boarding systems, to election IDs, to opening bank accounts, to buying mobile connections, to marriage registrations, would allow the Indian government to create a massive surveillance database akin to the Social Registry System.
The privacy experts were right — as noted earlier, the UIDAI was part of an Inter-Ministry panel on implementing the Social Registry System by June 2017.
But in July 2017, the UIDAI filed an affidavit in the Supreme Court claiming the opposite.
“By design, the technology architecture of UIDAI precludes even the possibility of profiling individuals for tracking their activities,” the authority stated in the affidavit. Dismissing the concerns of privacy advocates, the UIDAI falsely claimed that “contention of the petitioners that on the basis of a single identifier, Aadhaar will enable the government agencies to track and profile and do surveillance is completely unfounded and denied.”
“The biggest error apparent that the Supreme Court made in the Aadhaar case is it focused on what UIDAI can or cannot do with Aadhaar and ignored entirely what the Government can or cannot do with it,” said Anupam Saraph, a privacy advocate and a petitioner in the Aadhaar case.
Saraph said 32 separate petitions had been filed against Aadhaar, highlighting its potential for misuse in everything from money laundering to profiling individuals.
“The Supreme Court converted all of them into one petition against the UIDAI and decided the case on the basis of what UIDAI said it could or could not do,” Saraph said. “All those 32 cases must be reopened.”
In September 2018, the Supreme Court finally gave its judgment, restricting Aadhaar’s use cases to only distributing government’s subsidies to the poor.
The following year, in April 2019, the Ministry of Rural Development constituted an “Overseeing Team” to devise pilot exercises for updating the SECC data and “exploring alternative options to Aadhaar as a single identifier across programs”, the documents show.
Eventually, it was decided to simply change the Aadhaar Act instead. In June 2019, the Ministry for Rural Development prepared a note detailing the changes required to the regulations governing Aadhaar authentication, and the sharing of Aadhaar data.
On October 4, 2019, in the fourth and latest meeting of the inter-ministerial expert committee, the UIDAI said it was finalizing the amendments to the Aadhaar (Authentication) Regulations, 2016 and the Aadhaar (Sharing of Information) Regulations, 2016.
If implemented, these changes will render the Supreme Court judgment meaningless and even dilute the original Aadhaar rules: one of the key safeguards in the current rules states that the information received from a person in an Aadhaar-enabled transaction can only be used for the specific purpose for which their Aadhaar information was sought. This information cannot be shared further.
Other safeguards relate to the sharing of Aadhaar-authentication logs. These logs contain data on location and time of the authentication, apart from the demographic and biometric information of the Aadhaar holder. At present, the rules forbid an authenticating agency from sharing authentication logs with anyone apart from the holder of a particular Aadhar number.
But the Social Registry is premised on the notion that citizen information — particularly Aadhaar-related information — shared with any government department can be seamlessly incorporated into a unified database accessible to a third government agency. These safeguards, the officials at the October 2019 meeting complained, would make it very difficult to build the Social Registry.
One of the solutions proposed by the Ministry of Rural Development was that if a user consents to share their data with one government agency, it shall be assumed that they have consented to share their data with all government agencies.
As per the minutes of the October 2019 meeting, the UIDAI said, “all the amendments are in the pipeline.”
The “blanket consent without a finite and legitimate purpose” envisaged by these changes would be contrary to the directions of the Supreme Court, said Prasanna S., a privacy lawyer who argued in the Aadhaar case. Any state action that impinges on a citizen’s privacy, the court said in the Puttaswamy judgment, must pass the following test:
“The test mandates, 1. That State Action should be lawful - i.e. with the backing of enacted law. 2. It must pursue a legitimate purpose. 3. It must have rational connection to the object sought to be achieved by the law under which such state action happens. 4. Such state action must be necessary to achieve such objective. And 5. Such state action must be proportionate i.e. the least restrictive of all means available to achieve such objective,” Prasanna said.
“When it says it must be for legitimate purpose, it pre-supposes that such a purpose is singular and definite. Collection of data for one purpose and use of it for another purpose without consent or a consent for general purpose data-sharing are incompatible with the purpose limitation principle,” he added.
Tick tock to 2022
As things stand, the Modi government remains deeply invested in the Social Registry. Manoranjan Kumar, the man who first penned the foundational note that kicked off the process, has retired but his successors in bureaucracy continue to support the Social Registry in the slogans of the Modi regime.
“This Social Registry Information System strengthened upon SECC data, would go a long way to establish a much needed continuous, authentic, efficient, and evidence-based data management system in the country to identify beneficiaries of government welfare schemes and would be an enabler for achieving goals of NEW INDIA@2022,” Biswajit Banerjee, the joint secretary in the ministry of rural development, wrote in the file on June 17, 2019, referring to the government’s latest buzzword to mark 75 years of Indian independence.