India Launches First Population Census in Over 15 Years, Covering Estimated Over 1.4 Billion People, Caste Statistics as Core
India has begun its longest gap national population census since independence. The last complete census was in 2011, and it was postponed from 2021 due to COVID-19 and elections, creating a gap of about 15 to 16 years. The government claims this is the largest population count in the world, covering an estimated over 1.4 billion people, about 640,000 villages, and nearly 10,000 towns.
The cabinet approved a budget of approximately 1,171.824 billion rupees, equivalent to about 1.2 to 1.25 billion USD. Over 3 million census enumerators, supervisors, and grassroots officials will register households door-to-door, with about 3.1 to 3.3 million on the front lines. The first phase of housing and housing listing will run from April to September 2026, with continuous operations for 30 days in each state; the second phase of population registration is scheduled for February 2027, with the Himalayan snow region starting early in August to September. The reference date for most areas is March 1, 2027, and for the snow region, it is October 1, 2026.
This is India's first primarily digital census. Enumerators will directly upload data using a mobile application, supporting 16 regional languages, and can work offline before syncing with the server; households can fill in their information online first and receive a self-registration number, which will then be verified by field staff. The questionnaire has expanded from 29 items in 2011 to 40 items, adding fields for digital literacy, parental information, and optional fields for ID numbers, phone numbers, voter ID, passport, driver's license, and bank account numbers. Some sensitive defense areas in Ladakh will still use paper forms.
Caste statistics are a political core. The Political Affairs Cabinet Committee decided on April 30, 2025, to comprehensively record castes during the population registration phase, marking the first systematic enumeration of all castes in a decennial census since 1931, rather than just listing scheduled castes and tribes. Enumerators will fill in the information as stated by respondents without providing a preset list, and respondents can choose not to declare. The 2011 socio-economic caste census allowed open entries, resulting in about 4.67 million different caste names, while only 4,147 were recorded after the 1931 census; the original data from that census was not fully disclosed later.
The census results will reshape welfare distribution, urban planning, employment, and immigration scenarios, and will serve as the basis for the reallocation of seats in the Lok Sabha. The population in 2011 was 1.21 billion, and the UN estimates it has now exceeded 1.4 billion, making India the most populous country in the world. Complete data is planned to be released around March 2027 and will be provided to ministries in machine-readable format as "Census as a Service." The housing phase progress in West Bengal, Assam, Manipur, and other areas lags behind most states.
In market mechanisms, this is a state-mandated collection, not a free transaction. The buyers are the central and state finances, as well as the political machinery relying on the new denominator for quotas, subsidies, and electoral districts; the sellers are 3 million field workers, mobile applications, and data center contractors. Funds flow from budget allocations to enumerator compensation—if the average compensation is set at 25,000 rupees for both phases, personnel costs alone could consume about 775 billion rupees. Beneficiaries include populous northern states with new population weights, caste-mobilizing political parties, and consumer research institutions; those under pressure include southern states with declining birth rates, fearing a redistribution of seats, and policy departments concerned that open entries may render caste data unusable for quotas.
Source: Public Information
ABAB AI Insight
The decennial population machinery in India has never been interrupted since independence until it was halted in 2021 by the pandemic. The Modi government has turned the restart into a two-piece set: comprehensive digitization, along with expanding caste enumeration from "only listed groups" to the entire population's oral accounts. After the opposition turned the caste census into an electoral issue with the slogan "as many people as there are, there are rights," the ruling coalition shifted to acceptance before the elections in Bihar. This is not just a technical upgrade; it is about handing back the denominator that has been dormant for 15 years to the political parties for contention.
The capital path is straightforward. The largest portion of the 1,171.824 billion rupees is the foot compensation for 3 million people, followed by mobile phones, portals, monitoring dashboards, and publicity. The central government intends to harvest data as "Census as a Service," aiming to bypass the state-level statistical intermediary and deliver village-level housing, water, internet, and caste data directly to ministries. For investors, this is a new blueprint for consumption and labor; for the Ministry of Finance, it is a targeting tool for subsidies; for the Ministry of Home Affairs, it is fuel for redistricting. Open entry for castes means first collecting the raw noise, then deciding whether to clean it into a usable quota list.
Comparable examples include the U.S. decennial census determining House seats, and the allocation wars triggered in Brazil and South Africa after incorporating race or color into public statistics. India is more unique: after 1931, castes were removed from the census, suppressing identity politics with "equality" in the early years of independence; bringing them back in 2027 reflects that quota politics has grown too large to ignore other disadvantaged groups. The current phase is "data collection" rather than "data governance"—first getting 1.4 billion people into mobile phones, then debating how to merge 4.67 million caste names.
The structural judgment belongs to the transfer of pricing power. Whoever owns the new population and caste list will possess the pricing power for seats, reserved quotas, and welfare formulas for the next decade. The mechanism is that the outdated 2011 denominator prevents the population growth in the north from translating into parliamentary weight, and allows caste quotas to continue using old lists without new evidence. Digitization accelerates collection but does not guarantee faster publication; the detailed immigration table from 2011 took eight years to be released. The real power lies not in the moment of door-to-door enumeration, but in who has the authority to merge caste entries and who has the authority to delay publication.
ABAB News · Cognitive Laws
The hardest power of the state is deciding who is counted in the denominator
Digitization accelerates collection but does not automatically accelerate publication
Once identity is named, quotas will come to collect debts