I wrote earlier about how Census 2027 could lead to the unfreezing of Lok Sabha and assembly constituencies. This column looks at why they were frozen in the first place in 1971, and why these figures became sacrosanct.
To understand this, one has to revisit the Malthusian fears that dominated development discourse in the 1960s. The narrative of an ever-expanding world population eating up the earth’s scarce resources found support in the United Nations, the World Bank, the International Monetary Fund, and powerful foundations such as the Rockefellers, Ford, and the Club of Rome.
Demographers like Ashish Bose, who infamously coined the term ‘Bimaru’ for the four Hindi-speaking states of Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh, argued that the unbridled growth of population in this region was the albatross gripping the neck of development. This argument was accepted by then Prime Minister Indira Gandhi and her cabinet. During the two years of the Emergency (1975-77), her son Sanjay Gandhi, though unelected, exercised real power on account of being a dominant part of the Prime Minister’s Household (PNH), which had overtaken the Prime Minister’s Secretariat as the most effective power centre in the country. His driving of mass sterilisations in this era are well documented.
However, it must be placed on record that the Planning Commission and the health ministry also toed this line.
Also Read: Census 2027 is special. It could undo the 1971 freeze on constituencies
Demographic transition
This brings us to the concept of demographic transition — a movement from high fertility and mortality to low fertility and mortality. During the first phase, this leads to a surge in population due to the lag between the two.
As historian Ravi K Mishra has shown in his magisterial book Demography, Representation, Delimitation: The North-South Divide in India (2025), if we study India’s demographic profile in the long durée — by looking at census figures for nine decades, from 1881 to 1971 — it was the south that had higher decadal growth rates of population until about 1971, after which the north Indian states began to grow faster than the south.
We learn that Kerala doubled its population by 1931, “rising from slightly less than 50 lakhs in 1881 to nearly a crore”. From 1931 to 1971, Kerala’s population doubled again and then began to slow down. Its share in India’s population grew from less than 2.5 per cent in 1881 to 3.89 per cent, and then began to decline after 1971, reaching 2.76 per cent in 2011, as it became the first Indian state to reach below-replacement fertility.
A similar pattern can be observed in Tamil Nadu as well, where the demographic advantage had already accrued by 1971, making claims of political disenfranchisement anything but genuine.
Thus, while Bihar’s share of the overall national population fell from 8.9 per cent in 1901 to 7.7 per cent in 1971, Kerala’s rose from about 2.7 per cent to 3.9 per cent over the same period. Data from the four decades between 1971 and 2011 shows that Bihar’s share stood at 8.6 per cent, while Kerala recorded a marginal increase to 2.8 per cent. But having said this, in both absolute and relative terms, the north’s demographic spike has certainly been more pronounced.
How states have fared in the last two decades will be known with statistical accuracy from the forthcoming Census of 2027. But it is clear that, across the country, the share of the 0-14 age group has declined from 41.2 per cent in 1971 to 24.8 per cent in 2021, a clear indicator of falling birth rates.
Over the same period, the working-age population (15-59 years) has increased from 53.4 per cent to 66.2 per cent, suggesting that the country is in the middle of a demographic dividend phase, where the ratio of the working-age population exceeds that of dependents.
Also Read: Delimitation freeze gave us time, but now India’s democracy faces a ticking time bomb
Census of 2027: The nation talks to itself
The decision to conduct Census 2027 (the 16th census of the country and the eighth after Independence), and to include caste as a category of enumeration, was taken by the Cabinet Committee on Political Affairs on 30 April 2025.
Needless to say, the current census will be the biggest source of primary demographic data at the village, town, and ward levels, providing authentic information on various parameters, including housing conditions; amenities and assets; religion; caste; language; literacy and education; economic activity; migration; and fertility.
There are two phases of census operations. It will start with the enumeration of all buildings between April and September 2026. This will include every possible habitation, from the palaces of erstwhile maharajas to bungalows in Lutyens’ Delhi, housing apartments, EWS colonies, and those staying in JJ clusters, informal settlements, shelters for pavement dwellers, prisons, juvenile centres, and child-care homes. There could be more than one household living at the same address. Comprehensive information across 34 parameters— from the status of the roof (cemented or thatched) to water supply, metered electricity connections, latrines, and soak pits — will be collected for each family. This will provide the best possible data set for both urban housing and rural development departments.
The actual count of people will be carried out from 9-28 February 2027. Enumerators will take a count of people based on where they are located, except in snow-bound areas of Leh and Himachal. This poses its own challenges, as the period will also coincide with the Kumbh, when millions congregate in Haridwar. In this specific case, people will have the option of being counted in Haridwar or back home, where the head of the household will be asked to furnish information about each family member. Enumerators will collect details on age, gender, fertility, education, income, migration (for work or marriage), religion, profession (including unemployed status), languages known (mother tongue as well as languages spoken and understood) and caste status (SC, ST, OBC, or general).
Now that the census is digital, it will be possible to furnish this at the national, state, district, and sub-district levels. However, how much of it is shared, and when, and the level of disaggregation, is a matter of political discretion.
This largest administrative and statistical exercise in the world will cost the exchequer nearly Rs 12,000 crore. Over three million field functionaries will complete this gigantic exercise of national importance using a mobile app for data collection, linked to the Census Management and Monitoring System (CMMS), a dedicated portal developed to manage and monitor the entire process in real time. Ministries and departments have already identified the data sets most relevant to their policy needs, and Census-as-a-Service (CaaS) will deliver this data in a clean, machine-readable, and actionable format.
Parliament has been informed that the working results of Census 2027 will be available within a few weeks as all data is being captured digitally. However, many experts, including former Chief Election Commissioner S Y Quraishi, are sceptical. The ex-CEC has argued that, given the Delimitation Commission’s past track record, the government’s idea of conducting the 2029 elections on the basis of the 2027 Census is “mathematically impossible”.
In my next column, I will look at some of the issues, such as migration, that complicate the use of census data for political representation.
This is the second instalment of a four-part series on the census in India, based on a keynote address delivered at the annual juridical conclave at NUJS Kolkata.
Sanjeev Chopra is a former IAS officer and Festival Director of Valley of Words. Until recently, he was director, Lal Bahadur Shastri National Academy of Administration. He tweets @ChopraSanjeev. Views are personal.
(Edited by Asavari Singh)