Patralekha Chatterjee | Delay over Census hurting the way India is governed

We are past the halfway mark in 2023. By now, most eligible Indians have got their Covid-19 shots and India is on its way to recovery, as the Narendra Modi government proudly proclaims. There have also been Assembly elections held in several states. But we have not yet had the decennial Census. The last one was held in 2011.

India had conducted a census even during the Second World War. But this giant exercise, which began in 1881, had to be postponed in 2020 due to the Covid-19 pandemic. Arguably, the coronavirus pandemic turned lives upside down across the world. That India’s decennial census got hit is unsurprising. What is surprising is that three years down the line, we still have not had the time to hold a census. Bangladesh’s last census was in 2022. The same as Brazil. The United States and China had their census in 2020, despite challenging conditions.

Then, there are countries which have not had a census in decades, like Afghanistan, which conducted its last census in 1979.

There have been anxious murmurs about the inordinate delay in conducting the next census for quite some time, but it is only last week that it became very clear that it is not going to happen till after the Lok Sabha elections next year. The Registrar General of India’s recent order states that the deadline for freezing administrative boundaries has been extended to January 1, 2024, and a census can be carried out only three months after the freezing of boundaries of districts, tehsils, talukas and police stations. The census, as we know, is carried out in two phases -- the housing census and population enumeration. The data that comes out is critical for evidence-based policy-making and to assess the socio-economic progress in the country.

What does it mean to have no recent census data?

First, we must recognise that the importance of the decennial census stems from its scale and its credibility. As Dr P.V. Ramesh, a physician, and a former national and international civil servant, puts it: “The population census is an important apolitical, anonymous, value-neutral data management exercise.”

It is a large-scale population survey, down to the village level, and gathers detailed information about the population, demographics, and socio-economic characteristics of the country. At a time when there are jousts over data manipulation, most people believe it is difficult to manipulate.

Why would a country dawdle over a census unless it is ravaged by war or conflict or is simply ill-equipped?

“Afghanistan, for example, has not had a census for years, despite efforts. Not having a census could also mean that one does not want to acknowledge the demographic changes that have political and economic implications,” says Dr Ramesh.

The population data is the basis for defining the boundaries of electoral constituencies of the state legislature as well as that of Parliament when the delimitation exercise is taken up. But though the data is apolitical, it has political ramifications. “Contemporary census data can bust or endorse popular narratives in the political discourse such as a huge increase of a particular group. The census also throws up data which gives one an idea about the situation relating to employment, literacy, status of women, fertility, migration, etc. If we do not have current census data, inconvenient truths and myths get perpetuated. Moreover, allocation of resources to the specific groups that most deserve public investment is unlikely to happen and get misdirected,” he argues.

Researchers I have spoken to in recent days underline the challenges of not having the latest census data. “If we would like to know how many children we have to immunise, where we should open anganwadis, or the number of priority households to be covered in the public distribution system, we need the census. The census is the only independent source of population data down to the lowest levels of administrative levels, such as villages, which includes population characteristics such as age, gender, household size, and so on,” says Dr Aashish Gupta, a demographer affiliated with the University of Oxford. Many critical government programmes, such as immunisation, anganwadi services, or schooling require age data. “In the absence of census data,” Dr Gupta points out, “we are using administrative data, such as that from HMIS (Health Management Information System) for immunisation." However, as Dr Gupta notes, the administrative data is often dependent on who is reached by the government machinery and often has inherent biases. “These data are filled by government functionaries in the process of carrying out their routine tasks, and can be biased if people aren’t able to access these services in the first place. This makes an independent and complete census essential to public policy.”

Take another sector -- transport. “In 2011, for the first time, the census in India reported travel distances and how people travel to work. This did not include agricultural workers but is very significant. The same question was asked across the country and the census threw up information which we never had earlier about travel patterns. For instance, it told us that women in India do not use cycles and motorcycles to travel to work much, and their dependence on walking and public transport was much higher than that of men. This of course has a bearing on the choices open to Indian women,” says Rahul Goel, assistant professor, Transportation Research and Injury Prevention Centre, Indian Institute of Technology, Delhi.

Researchers face daunting challenges since the official data from India on this front, as in many other areas, is very old. There is no updated census data on travel patterns. “Currently, we use vehicle registration data, but that offers a partial picture,” says Mr Goel. In India, many use rickshaws or tractors to travel from one point to another. This is not captured in any current data. “The population data, as in the census, is important for us because of the granularity it offers. Currently, even for estimating road death rates, we only have population projections at the state level,” he points out.

We do not know how cities have grown, so we do not really know the road traffic death rate (per 100,000 population). And this is just another addition to the long list of the things “we don’t know”.

Researchers do the best they can. But there is an “issue of comparability if different groups use different sources for their population estimates”, says Mr Goel. If we are using projections by independent agencies, we need to mention that.

Which brings us back to the key issue: without recent census data, it is difficult to accurately gauge where the needs are the most.