By M Rajshekhar, ET Bureau | 10 Sep, 2013, 05.31AM IST
In the leaky system of welfare delivery, databases are the newest valve that governments are installing to ensure that benefits reach those—and only those —they are intended for.
Since December 2012, for instance, the government of Madhya Pradesh has been appending on to the Centre's Socio Economic and Caste Census a host of household-level data: bank account numbers, NREGA card numbers, welfare entitlements, land ownership, whether their house is kuchcha or pucca, if it has a toilet, and so on.
In 2008, the Congress government in Delhi had similar intentions in mind when it decided to overhaul its welfare delivery. Today, that failed experiment with databases reveals everything that can go wrong with databases: from exclusion to selective updation, from political profiling to privacy issues.
So far, about 85% of households in MP, or 25 million people, have been surveyed, says Aruna Sharma, the state bureaucrat driving the programme. MP intends to use this database to determine eligibility among the poor for welfare entitlements like pensions, scholarships and food, with the emphasis being on accuracy.
"We can run queries like, 'how many women live alone in kutcha houses in rural MP?', and start a widow pension for them after field verification," says Sharma, secretary for rural development, panchayati raj and social justice. Yet, as the example of Mission Convergence in Delhi shows, even a supposedly objective system can be ruined by subjective choices and neglect.
In 2008, the Congress government in Delhi had similar intentions in mind when it decided to overhaul its welfare delivery. Today, that failed experiment with databases reveals everything that can go wrong with databases: from exclusion to selective updation, from political profiling to privacy issues.
Good intentions
In 2008, despite running several welfare programmes, Delhi was faring poorly on Human Development Indices. The existing system of targeting was not working.
"The only people receiving benefits were those known to the department or the politicians," says Rakesh Mehta, who was Delhi's chief secretary at the time. This was partly because there was no accurate way to identify the poor.
Databases of welfare departments were unreliable; no census had been done to ascertain BPL (below poverty line) households; every welfare department followed a different yardstick to gauge eligibility.
"Poor poverty estimates resulted in low budgetary estimates on welfare outlays. This, in turn, resulted in further exclusion," says Amod Kumar, head of the community health department at St Stephen's Hospital, who worked on this overhaul in its early days. In response, the Sheila Dikshit government took a two-pronged approach.
One, like MP, it decided to created a database of all vulnerable households in the state. Two, it outsourced the last mile to NGOs. 'Gender resource centres', each managed by an NGO, were made responsible for a cluster households, of 10,000 to 100,000.
They had to keep this database updated, inform people about new schemes and help them access entitlements by functioning as a single window to government services. In the new system, selection of beneficiaries would move away from departments and local authorities like MPs and MLAs.
Instead, the database would be queried for names of households/individuals eligible for welfare schemes. Five years on, Mission Convergence and its database-driven approach lies in shambles. Women ET spoke to complained of exclusion from the state food programme and a lack of awareness about government programmes.
Some schemes like pensions have gone back to their mother departments. Fieldworkers complain about low salaries and overwork. In areas where GRCs have quit or are not working well, locals have no recourse.
Poor outcomes
Kumar says Mission Convergence faced challenges at every step, starting with the survey. It was stopped after covering 60% of the city as the budget earmarked by the Delhi government was reached. As a result, it under-reports the number of beneficiaries in Delhi. Take Annashree, the Delhi government's cash-in-lieu-of-food programme.
"There are 200,000 beneficiaries (drawn from the database) in a state which has many more who deserve it," says Biraj Patnaik, a right to food activist. Patnaik attributes this to Delhi's "anti-migrant" stance. In Delhi, only those who have stayed in the city for at least three years are eligible for benefits.
Delhi chief minister Sheila Dikshit defends this filter. "About 65% of Delhi's budget is spent on welfare," she says. "There are cases where people living in Ghaziabad have shown themselves as residents of Delhi to claim benefits." The next challenge came from the establishment. "There was a stormy meeting between Dikshit and the MLAs," says Kumar.
"They asked questions like, 'once we have certified eligibility, how can an NGO sit in judgement?' Eventually, a compromise was worked out. People whose names were thrown up by the database would also have to get their form signed by the MLA." The MLAs and their discretionary powers persisted in the process.
The survey, done in 2008, was not repeated. Selective updation happens if people ask for it or come to a GRC for Aadhaar enrolment. "The purity of the data was the critical element," says Mehta. "This was to be updated every six months and a more detailed survey was to be done every four to five years. This updation has not taken place."
This problem is compounded as urban populations are mobile. Take Ghevra, a resettlement colony in north-west Delhi housing those displaced by slum demolitions in Delhi.
People started settling here around 2007. But since their new houses were far from their places of work, most families —three out of four, say residents —have relocated. Newer families have taken their place. Between high mobility and low updation, several households are missing from the database.
"The database is not outdated," defends Nila Mohanan, mission director of Mission Convergence. "In 80% of the cases, we find people there." However, numbers given by a community mobiliser at the GRC in a South Delhi locality for Annashree suggest the mismatch is greater.
Recently, the database was used to generate names of those eligible for Annashree in the locality. "We were given 2,442 forms (generated by the database)," he says, not wanting to be named.
"Of this, we found that 1,100 households had left the area. Another 700 or so were ineligible. Another 3,000 families are eligible, but are not on the database." Counters Mohanan: "Whenever we find a person is not at the address listed, we try and reach them through their phone number listed. We also ask people to come on their own and update their entries."
But she agrees a fresh survey is needed. Chief minister Dikshit feels the database does not need to stay updated. "In Annashree, we did the first round of enrolment based on the database," she says. "But in the second one, we are allowing others to apply as well." Besides subjectivity, there are concerns around privacy.
In Mission Convergence, the GRC, as well as the district resource centre (which oversees the GRCs), has access to all data about locals in their clusters. "Data cannot be shared without prior permission," says Mohanan. But, in the above-mentioned South Delhi locality, ET was given a copy of a completed Annashree form for a household.
It included queries on whether anyone in the household had TB or AIDS. "There can be no sharing of data with other departments or organisations without the informed consent of the person," says legal researcher Usha Ramanathan. Her other concern is that the state is collecting more information about people than it needs.
"The first rule of data protection is the 'collection limitation principle'," she says. "You do not collect information that is unconnected with the purpose for which the data is being collected."
Adds a senior official of the Registrar General of India, on the condition of anonymity: "Any database-style governance must have a privacy law underpinning it. Where's that law?" All relevant questions for MP and other states that embrace databases for welfare delivery.
