India’s 2027 Census demands careful planning to get outcomes right

The projected total population figures from different national and international organizations differ so substantially that they can turn any development narrative upside down. 

The Prime Minister’s Economic Advisory Council had raised the issue that using India’s urban population share derived from the National Sample Survey (used in the absence of Census data) tends to underplay national achievements. This is because development indicators for urban areas, which get a lower population weight even though urbanization must have increased, are generally higher than those for rural areas. The more progress India makes, the greater the under-estimation. 

Also Read: Sanjoy Chakravorty: A caste census is a Pandora’s Box that India must open anyway

Unfortunately, a projection of India’s urban population by an expert group under the ministry of health did not help, since it was based on the urban growth figure of the period 2001 to 2011, while the subsequent trend rate is likely to have risen. Also, the claimed achievements under national schemes such as the Jal Jeevan Mission, Swachh Bharat and others are based on administrative reporting and are thus expected to be high. New Census data on housing amenities can validate some of these claims, as questions on a household’s source of drinking water or use of toilets are generally asked as part of the questionnaire. 

Another debate is on India’s civil registration data showing a very big jump in death registrations in 2021. Attributed to covid, this number contradicts the official claim on such deaths. Given the limitations of our civil registration system, the forthcoming Census could add a few additional questions to validate covid mortality data.

In conducting the Census, it is important to revisit the rural-urban segregation of settlements. Urban boundaries along with the division of districts have already been frozen. The usual practice at the state-level directorates of census operations is to identify census towns in advance by using demographic data from the previous Census (the 2011 headcount in this case). But this will lead to an unrealistic assessment of urbanization based on data that is a decade-and-a-half old. The Registrar General could consider declaring census towns after the 2027 Census by using its results for a realistic rural-urban division.

Also Read: Himanshu: India’s caste census must serve its purpose

Updated Census data would be of great use to local governance agencies like gram and block panchayats (most of which still show 2011 census data on their websites) for projecting resource needs. Survey agencies are also eager to use current population data for sampling purposes. 

A major challenge is the inclusion of the caste question in the Census. The Socio-Economic Caste Census (SECC) of 2012 was conducted outside the scope of the Census Act of 1948, since it involved validation of responses from households at the ward or gram sabha level, which would have violated the Act’s confidentiality clause. 

The robustness of state-level caste survey data has been questioned on doubts over the expertise of the agencies involved and their method of canvassing information. The SECC caste data had to be dumped as the list of castes was not predetermined and the total number of groups ran into hundreds of thousands.  People understood the questions differently and often reported their occupation as caste, since this choice was left to respondents. 

The same caste reported in English and various local languages and dialects compounded the problem. To overcome such problems, an acceptable list of castes for the whole population must now be prepared beforehand.

Also Read: Caste census? Okay, but we must handle it with care

The approach taken to record the caste status of Scheduled Caste and Scheduled Tribe populations is always state-specific. This can be adopted in preparing lists for Other Backward Classes (OBCs) too. A person who identifies as OBC should select her caste from a central list or that of the state where she lives. However, state-level caste lists must mention all existing groups in that state and be acceptable to all communities. Minor spelling or  linguistic variations must not result in exclusion. 

Understandably, it would be a challenge to prepare central and state lists. Significant inter-state migration makes the task that much harder, while the emergence of caste data from the Census could open up space for political bargaining through party politics and new norms of power sharing. 

Sundry answers sought by the Census may also require a relook. In the past, the census office has valiantly tried to gather information considered beyond its field of competence: on economic activity, occupation or physical disability, for example. It also sought answers on the possession of items like a radio, telephone, television and cycle. With substantial progress in terms of household amenities, electrification, communication and cooking gas connections, one can question the need to expend resources on collecting such details. 

Also Read: Census call: India’s emergence must get its balancing act right

The delimitation of electoral constituencies for the Lok Sabha and state legislative assemblies is to be based on population data from a Census conducted after 2026. The 2027 exercise could therefore be the basis for redrawn constituencies. Low-fertility states have been pressing for a formula that fixes the number of parliamentary seats based on some principle of ‘federal justice’ rather than demographic weight; this could become a point of political friction. 

Several states have also urged the 16th Finance Commission to increase the weight assigned to population from 15% to 22.5% for its fund-devolution formula, arguing that the 2011 Census data is outdated. No new announcement has been made by the government on using Census data to update the National Population Register or prepare a National Register of Citizens, but the latter could potentially be another source of controversy. 

Also Read: Himanshu: India’s economy is too complex to afford less-than-robust statistics

We need to proceed with care. The next Census demands a balanced and transparent approach. This applies to every aspect, from designing the questionnaire on the basis of a consensus to the methodology adopted for caste enumeration. It will be the most keenly watched data gathering exercise anywhere in the world. 

The authors are, respectively, former acting chairperson of the National Statistical Commission, and senior advisor at Development Alternatives.