The Government of India has announced that the next Census will be conducted in two phases over 2026 and 2027. The effective date for which it will provide all the data will be March 1, 2027.
India has undergone a Census once every decade since 1881; regardless of who ruled over the country and any calamity. However, the link was broken when India failed to conduct the Census for 2021 in the wake of the Covid pandemic. It is noteworthy that the delay in India’s case — 6 years — is much longer than in most other countries (including India’s neighbours such as Pakistan, Bangladesh and Sri Lanka).
Explaining the delay, the Ministry of Home Affairs (MHA) spokesperson posted the following on X: “Covid disrupted all sectors, including education. Around 30 lakh enumerators are needed for the Census and most of them are primary school teachers. Conducting the Census post-Covid could have disrupted primary education immensely.”
A concern for data quality
Most of the focus now is on two politically sensitive issues.
One, the caste census, or the collection of caste data along with all the other data. Two, the possible usage of Census 2027 data for the purposes of delimitation, which is essentially an exercise that will entail rebalancing the state-wise share of Lok Sabha seats based on population. Most of the southern states, which were more effective in population control over the past 50 years, stand to lose electoral representation relative to northern states, which failed to control their population growth as effectively.
At one level, the fact that caste and delimitation are heavily politicised, warn some demographers, may vitiate the quality and credibility of the Census data. This may happen as people attempt to tweak their answers on questions (such as caste and language etc.) in a manner that, in their view, suits their politics instead of providing an unbiased response capturing the reality.
Regardless, the Census’ importance goes far beyond these issues, and its data will be of critical importance for the functioning of the Indian economy.
Census’ importance for the economy and policymaking
The national Census provides the fundamental statistical basis for all other types of surveys and analysis in the country. The Census isn’t just the complete enumeration of all Indians. Rather, it is as complete an image of India as possible, as it collects data on age, gender, family status, economic status, unemployment, language, education, disability, migration — the list is pretty long.
In other words, the Census is a reality check for the country — a look in the mirror without which a self-portrait is not possible. In providing all this information for each Indian, the Census not only tells the story of how India changed in the past 10 years — in this case, the past 16 years — but also gives clues to how it might change in the next decade.
The Census forms the basis of all future surveys and policymaking. Here are a few examples.
1. Inflation control and Interest rates
For instance, how does the Monetary Policy Committee of the Reserve Bank of India know whether to raise interest rates or cut them or to maintain the status quo?
The RBI, of course, bases its decision on the retail inflation rate in the country. But how does the country know what is the retail inflation rate?
To be sure, retail inflation rate captures the rate at which the general price level in the economy went up from one year to the next. This rate, in turn, is based on the Consumer Price Index (CPI) and how much it rose over the past year. The CPI is essentially a basket of goods and services that retail consumers “consume”. The index allocates different weights to different goods and services and then maps how prices of these components move in a weighted fashion. Food articles, for instance, have a weight of 46% in the total index. This is another way of saying that 46% of all the money spent by a common Indian is spent on food items.
But how do we know what weight to allocate to food items while ascertaining the overall inflation rate? That, in turn, is based on a survey — a consumption survey done by a government agency.
But how does the government agency know whether its modelling of the survey, or the architecture of its survey, truly captures the reality?
It is here that the Census matters because it is the Census that establishes the “reality” that all other surveys try to capture. The Census provides the base for all surveys and policymaking. A survey is effective if it can capture the reality — the reality as depicted by the Census. Who lives where, how much does he or she earn, are they part of a family, are they migrating for work, where do they live etc.
A Census — say Census 2011 — provides one clear image of India at one specific point in time. That distribution of people — quality of their lives, their economic status, their migration patterns etc. — provides the base for all future surveys such as the consumption survey for government or the many of the market surveys done by private companies trying to figure out whether there is going to be more demand for cars ( and if so, what kind of cars) in rural India in the next 5 to 10 years.
For argument’s sake, if over the past 16 years, an average Indian’s expenditure on food had fallen to only 30% of his total expenditure then by giving a higher weightage to food articles, India has been overestimating (and overstating) its food inflation as well as its overall inflation. By the same logic, Indian policymakers would have been making the mistake of running a much tighter monetary policy — read maintaining high interest rates — and thus unintentionally holding back India’s economic growth.
Food inflation is just one example. There are many other variables for which the Census is absolutely critical. Take, for example, issues like migration and urbanisation.
2. Migration, Urbanisation and Governance
Migration, for instance, is one of the least understood variables. Often, population projections are based on fertility (birth rate) and mortality (death rate) assumptions. But there is a third variable that can, and often does, change the population of a village, city or state — that’s internal migration in the country.
The popular perception is that most of the migration happens from one state to another. But the fact is that inter-state migration is barely 12% of total internal migration in India. Most of India’s migration is essentially “local” — intra-district (62%) and inter-district (26%). Barely 4% of India’s total population participates in inter-state migration and this is hugely responsible for India having one of the lowest levels of internal migration in the world.
What’s more, most internal migration is between one rural area to another (almost 48%). The second biggest category — and also the fastest growing one — is from one urban area to another urban area.
As revealing as these statistics are — the fact is that they are based on 2011 Census data. The truth about India’s internal migration might have undergone a complete change since then. Only a nationwide Census can capture how reality has shifted.
Migration data has wide ranging policy implications — from affecting Budget allocations for rural and urban areas to policies regarding education, employment and social welfare and, as more and more cases of language divide come up, of social cohesion.
A necessary corollary is the impact on urbanisation. There is no clarity on how urbanised India is. Estimates range from 30% to 70%, depending on what definition one chooses for an urban area. Accurate knowledge of urbanisation is critical to India’s GDP growth, after all, according to a Niti Aayog estimate, 60% of India’s GDP comes from Indian cities, and they account for just 3% of the nation’s land.
If more people reside and work in cities, it should ideally lead to a fundamental change in the way Indian cities are governed, both in terms of how they taxes and what services they receive in return.
No alternatives to the Census
As explained above, no matter how good an agency it may be, no survey will be able to capture the reality if it is modelled on outdated Census data. This holds true for both public and private sector surveys. In fact, there is an element of rising costs of accuracy. With each passing year, it becomes more and more costly to get an accurate picture of reality.
Further, while it is true that there is an increasing volume of administrative data (that is, the data collected by different state and Union government departments) on a whole host of variables, from water and electricity availability to education and disability, the fact is that such data is often a very poor and, more worryingly, a very faulty and misleading substitute.
For one, there is a problem of comparability when looking at administrative data from one department to another, and even more so from one state to another. That’s because of differences in definition or the way data is collected etc. Not to mention the rather obvious problem: Data collected by the departments (and the officials) who stand to be adversely affected if it shows less than the advertised success is often unreliable. A glaring example of this came out when the latest round of the National Family Health Survey in 2020-21 showed that, contrary to the government’s claims that India had become Open Defecation Free, 30% of the households surveyed did not have toilets. That was just a survey, not a Census.
Upshot
The shifting of timelines — as a result of the six-year delay — as well as the politicisation of several issues that will be mapped are likely to create problems of analysis and comparison when Census 2027 eventually happens. However, it is also true that with every passing year of delay, India and its policymakers were progressively failing to get a clear picture of India’s actual reality. To that extent, this is a welcome development and will provide a critical reset not just for India’s economy and its policymakers but also its citizens.
Share your views and queries at udit.misra@expressindia.com
Take care,
Udit
The shifting of timelines — as a result of the six-year delay — as well as the politicisation of several issues that will be mapped are likely to create problems of analysis and comparison when Census 2027 eventually happens. (Express photo by Gurmeet Singh) 
Share
