The contemporary urbanisation paradigm in India is rooted in a visible growth of urban areas beyond large cities, a trend that is distinctly evident since the 2011 census. A lot of these urban areas are only defined but not governed as urban, and are known as census towns (CTs). Beyond summing up salient features of these towns, this paper tries to highlight the future trajectory of these areas, in order to outline what factors will drive these areas in future, and how policymakers and the state will respond to various demands of these places. Some important findings of this paper are highlighted below, through a short discussion with the authors.

In this paper, you have tried to predict the number of census towns for the upcoming census. How can you do such an exercise for the 2021 census now, in 2018?

In order to answer the question, let us look at the different steps that are involved in the classification of rural and urban areas in a census. The Registrar General of India (RGI) actually uses data from the last census for making such classifications. For example, the 2001 settlement data was used as the reference data for the 2011 census. The first step is to take note of the jurisdictional changes to villages and statutory towns (STs) that occurred in between two censuses. Statutory towns are urban areas established under a state or central law and are governed by urban local bodies. Since between 2001 and 2011, state governments formed new STs, and geographical area of the existing STs also underwent changes (mostly increased), it is important to adjust these changes for the 2011 census data. Similarly, new villages also came up during this period.

Once the list of STs is finalised, the second step involves identification of census towns (CTs). For that, the RGI takes the data from the last census (i.e., 2001 data for 2011 Census), and sees how many villages satisfy three pre-defined conditions to become CTs. The three conditions are: (i). population of 4,000 and more, (ii). population density of at least 400 persons per square kilometer and (iii). male non-farm workforce of more than 75 percent. Note that the population used is 4000 and not 5000, as in the definition of an urban area. This is because the RGI presumes the reduced population cut-off of 4000 should increase to 5000 during the intercensal period.

Since the previous census data is used for identification of CTs, one can use the existing 2011 census data to predict the number of CTs for the upcoming 2021 census.

Why does the Census make such classification before the census?

The tradition of identifying CTs before the census has been followed since the 1961 census. It is important to ask whether there are enough reasons to support the current approach of identifying CTs using previous census data instead of identifying CTs after the Census, using the actual census data. In our paper, we have shown how this method can lead to misclassification of CTs; i.e. 736 villages which were identified as CTs do not fulfill the criteria in the actual data and similarly 1400 villages actually fulfill the criteria but were not identified as CTs.

On the other hand, the current approach could be supported on two angles. Firstly, the set of information collected from the rural and urban areas is not exactly same. There is some information which is collected specifically from rural areas (such as information related to land use, irrigation, roads, Public Distribution System, nutritional and child care facilities etc.) and other information which is only collected from urban areas. As a result, classification of a settlement into either rural or urban has to be done before the census so appropriate information can be collected. Secondly, finalising rural and urban frame before the census also helps in releasing the rural and urban population data much faster.

What are the challenges in such estimation?

The main challenge in estimating the number of CTs from the census data is due to the paucity of information in the census data. The census manual indicates that the workers who belong to the “Plantation, Livestock, Forestry, Fisheries, Hunting and allied activities” (PLFFH) sector must be treated as farm employment in the identification of CTs. But, since the 2001 census, village level workforce data is available in four broad groups where PLFFH workers are clubbed in a group that is primarily a non-farm sector. As a result, some type of adjustment is required and there are limitations to it.

Besides identifying villages that you are expecting to become census towns in the coming 2021 census, are there any new findings in this paper?

This work is based on our previous work on census towns. A number of researchers from Centre for Policy Research (CPR) as well as our collaborators have worked on various aspects of CTs in the past. One important aspect that we have focused on in our previous work is the extent to which proximity to urban areas plays a role in the formation of CTs and its relation to their characteristics. In this paper, we have tried to delve deeper into the spatial characteristics of CTs. For example, even if a CT is not proximate to a large town it is important to know whether it is a standalone CT with a more local economic interaction or part of a cluster of CTs where agglomeration economies may come into play. Similar classifications can also be made for proximate CTs.

Are these census towns more like villages or like towns?

There can be different answers to the same question depending on what we are comparing. We can compare intensity of non-farm activities, economic prosperity of citizens, structure of the society or provision of public services etc. In this paper, we have used some indicators related to access to public services and private assets, as well as intensity of night-time light data (i.e. satellite image of the earth at night) to capture the levels of economic activity. For all these indicators, we found that CTs are better than villages of similar population size and for some indicators they are comparable with smaller STs.

What are the policy implications of such large number of upcoming CTs on the governance of these settlements?

It has been mentioned earlier that while CTs are governed like other villages in the country, they are counted as urban, and their economic characteristics are different from their rural counterparts. As mentioned, CTs in a large number are expected to be added to the already existing stock in the upcoming census. So, planned governance of these settlements is crucial for their sustained economic growth. While the current policy discourse does not offer anything specific regarding CTs, the Central Government in a recent advisory asked the states to consider converting them into STs. Other than the fact that such conversion is a variedly contested process, the diverse nature of CTs highlighted in this paper questions the ‘one-size-fits-all’ approach and prescribes a more integrated approach from the State.

The working paper can be accessed here.

Additional research on census towns at Centre for Policy Research can be accessed here.