India’s long-delayed census is finally in the field. Houselisting and housing operations began state by state on April 1 and will continue through September 2026, ahead of population enumeration in February 2027. For the first time, enumerators are collecting data through mobile applications, households have an option to enumerate themselves online, and the population count will include caste. The reference date for most of the country is March 1, 2027.
The digital machinery attracts attention, but the larger significance of Census 2027 comes from the gap it will close. India’s last complete population count was in 2011. Since then, the country has urbanised, migration patterns have shifted, household structures have changed and the economy has become substantially more formal and digitally connected. Yet a large part of policymaking still rests, directly or indirectly, on a demographic benchmark that is 15 years old.
READ | The caste census debate is really about citizenship
That makes the census part of India’s economic infrastructure. Roads and power networks connect markets physically. Statistical systems allow governments and businesses to see those markets, estimate demand and decide where scarce resources should go.
The cost of a 16-year data gap
A census does something that administrative databases and sample surveys cannot replicate. It establishes the population denominator and provides information at a geographic scale that surveys generally cannot reach. Population projections can bridge the years between censuses, but the longer the interval, the more room there is for assumptions to diverge from reality.
The consequences are visible in public policy. Under Section 9 of the National Food Security Act, the number of people to be covered by the targeted public distribution system is calculated using population estimates from the latest census whose figures have been published. The national coverage ceiling continues to be 81.35 crore people, a figure derived from Census 2011. The government reported that states and Union territories had identified 80.56 crore people against that ceiling as of February 2026.
The issue extends well beyond food subsidies. Schools, hospitals, transport systems and housing programmes depend on estimates of where people live and how populations are changing. Per-capita indicators can also become misleading when their denominators lag behind demographic change.
Businesses have a similar problem. Decisions about distribution networks, retail expansion, housing, logistics and consumer markets depend on information about household size, age composition, migration and urban growth. A credible census will provide a new benchmark against which surveys, projections and market estimates can be tested.
READ | Caste census is just first step — Govt must act on the data
Digital enumeration raises the stakes
Technology should shorten the journey from household enumeration to usable statistics. Census 2027 uses mobile applications for field collection, a central monitoring system and an online self-enumeration facility. The government says more than three million enumerators, supervisors and other officials will participate.
The design contains an important safeguard against digital exclusion. Self-enumeration is optional. Enumerators are still required to visit households, including those that have submitted information online, and confirm the data. This matters in a country where access to smartphones, connectivity and digital literacy remains uneven.
Digitisation also changes the risk profile of the census. Paper records created logistical problems; centralised digital collection creates cybersecurity and access-control risks on a different scale.
The legal protection for census information is unusually strong. Section 15 of the Census Act bars individual census records from inspection and from being admitted as evidence. The government has also said that census data will be stored in data centres designated as critical information infrastructure.
Those protections will matter because the census asks citizens to disclose information that they reasonably expect the state to keep confidential. Public confidence will depend on how securely the data are handled and how clearly the safeguards are explained.
Census 2027 will reshape political debates
The economic value of the census cannot be separated entirely from its political consequences. The 2027 exercise will collect caste data beyond the enumeration of Scheduled Castes and Scheduled Tribes for the first time since Independence. The government decided in April 2025 to include caste enumeration in the population phase.
That will bring a new empirical basis to arguments over representation, affirmative action and the distribution of public resources. The numbers themselves will not settle these disputes. They may well intensify them.
An even larger question concerns delimitation. Article 82 of the Constitution has protected the existing interstate allocation of Lok Sabha seats until publication of the relevant figures from the first census taken after 2026. Once Census 2027 figures are published, India enters a new constitutional phase in which parliamentary representation can be readjusted.
This has particular significance because population growth has varied considerably across states. Any future redistribution of seats will have to reconcile democratic representation with concerns in states that achieved lower fertility earlier. The census will provide the numbers around which that argument takes place; it cannot determine the political settlement.
The stakes are higher still because the constitutional provision for reserving one-third of Lok Sabha and state assembly seats for women is to take effect after a delimitation exercise following publication of figures from the first census conducted after the 2023 constitutional amendment.
Credibility therefore matters far beyond statistical administration. Errors, unexplained revisions or long delays in publication would quickly acquire political meaning.
READ | Caste and power: Census 2027 prepares for a historic reckoning
The value lies in what happens to the data
Completing enumeration is only one part of the job. India needs the results quickly enough for them to influence decisions.
That requires timely release of tables at useful geographic levels, transparent documentation of methodology and a clear publication calendar. Researchers need access that permits comparison with earlier censuses. State governments need figures that can be incorporated into planning. Businesses need reliable demographic benchmarks rather than another round of projections built on 2011.
The census will also expose places where policy assumptions have drifted furthest from reality. Some towns will turn out to have grown faster than expected. Migration corridors may look different. Population ageing will be more advanced in some states than national averages suggest, while others will still have large cohorts entering the workforce.
These are not statistical curiosities. They influence the location of hospitals, schools, transport links, housing and jobs.
India has spent much of the past decade building sophisticated systems for digital payments, tax administration and welfare delivery. Those systems produce enormous volumes of real-time administrative data. They still cannot answer the elementary questions for which a census exists: how many people live in a particular place, what their households look like and how the geography of the country’s population has changed.
Census 2027 can repair that weakness. Its success should eventually be measured by the credibility and usefulness of the data it produces, rather than the share of forms filled online. After a 16-year gap, the costliest outcome would be a census completed on schedule whose results arrive too late to change decisions.
Dr John Felix Raj is the Vice Chancellor, and Dr Sovik Mukherjee is Assistant Professor of Economics at Xavier Business School, St. Xavier’s University, Kolkata.