From the traditional belief that India lives in villages to the conviction that this is India’s urban century, the country’s idea of socio-economic growth and development has transformed significantly in the last 75 years. To understand the change in narrative and policy response, the processes of the rural-to-urban transition, its pace and measurement, and the challenges and opportunities it creates, Om Prakash Mathur, one of India’s senior-most policy advisors and scholars on urbanisation, spoke to Shivani Singh.
Mathur recently released Changing Paradigms of Urbanisation: India and Beyond, a collection of 20 papers written between 1981 and 2023, which were commissioned by international agencies such as the World Bank, UN affiliates, universities, and research institutes. The book covers a wide range of issues – from urban sustainability to regional planning, governance to local government finance, and rural-urban balance to the challenges of urban poverty and the informal sector.
Edited excerpts from the interview:
Is the definition used by the Census of India to classify urban areas (31% in the 2011 census) adequate to get the real measure of urbanisation in India?
Until the 1951 census, the cut-off for an urban settlement was a population of 5,000. Ashok Mitra, the then census commissioner, changed the definition in 1961, stating that urbanisation is not just a demographic phenomenon; it has a strong economic orientation and, likewise, a solid social base. He introduced the definition keeping the (population) cut-off at 5,000, the 400 persons per sq km density, and 75% of the male workforce (employed) in non-primary sector. We have maintained this definition since then. And I would support the continuation of this definition. We have not yet reached a stage at which we could eliminate either of the two additional criteria introduced in 1961.
Villages are also categorised by size, and the number of villages in the size category of 5,000 to 10,000 is pretty large. As and when they acquire a population density of 400 persons and (men here) take up non-primary sector activities, these villages will also become urban.
The problem is not how the census defines urban. It’s that state governments are not treating census towns as urban (which they can) by giving them statutory status. They continue to keep them as rural (under panchayats) for no reason other than our economic policies. They get more grants from the central government for being rural. If they become urban, all the grants they get will disappear.
However, there are other ways in which resources flow from the centre (and could be tapped). The 15th Finance Commission report gives a separate allocation for metropolitan areas on the condition that all the census towns within the metropolitan area’s periphery are to be included. So, many census towns are now being treated as urban to receive this grant. The National Urban Renewal Mission grant was also given to the metropolitan area, consisting of census towns, large villages, etc. However, many cities complained that the central city (state capital) would keep all the money. (While these are issues that need to be resolved) there is, at least, a recognition by the Finance Commission that such areas cannot be treated as rural. That is a significant gain.
It is argued that urbanisation levels would be much higher if we followed the definitions of other countries or used proxy indicators such as night-light data. Your comments?
The 2016 economic survey stated that if we adopted the Venezuelan definition, where the cut-off is 2,500 persons, India would have been 65% urban in 2011. But Japan uses 50,000 population as the cut-off point, and settlements below 50,000 are rural. If we take this definition, India would be only 10% urban.
Why are they not giving the example of China, where a city is (by definition) a settlement with a population of 100,000, and anything below that is not a city? Even then, China has become 55-58% urban within a very short time.
There is a lot of discussion on using night lights to measure urbanisation. But we must remember that India is a ‘one bulb economy’. Only urban areas have households with more than one bulb. Talking about the intensity of light, the 2016 report of the World Bank on urbanisation in South Asia shows the Delhi-Lahore route to be the brightest from a night light point of view, and implicitly, this route is more urban. So, will it be (possible to) apply the census definition to settlements along this route?
My position is, why use a measure that can often change in intensity? For example, households can start using two bulbs (over time). The census definition gives us a firm estimate and will do so over time.
There is a tendency to demonstrate that India is urbanising rapidly. Unfortunately, that is not the case. Our rate of growth is, at best, moderate. My paper shows that even by 2047, India will be just 51% (urban) unless the definition is changed.
Will changing the definition of urban (in the census) and including more areas improve the availability of services or infrastructure in India’s towns and cities?
If you change the definition by removing density and the occupation criteria, our urban areas would be the least productive in the world. The per capita GDP would remain the same. This means that a country with 50% of the population having a per capita GDP of USD 2,289 (in 2022) is the poorest country in the world.
While it is widely believed that urbanisation in India is primarily due to rural-to-urban migration, you point out that it isn’t so. What is driving urban growth now?
There are three sources of urban population growth. One is the natural increase. We are reaching the replacement level (the level of fertility at which a population exactly replaces itself from one generation to the next) in India, barring a few states. The southern states are below (the replacement level).
The second source is rural-to-urban migration. The third one is precisely what has happened now— incorporating rural areas into urban or rural settlements changing their economic and social character. This is what happened in Bengaluru and Hyderabad, where all the villages in the neighbouring areas were legally incorporated into the municipal limits over time. These are the ways in which urban population increases and that’s really what has been happening all through history.
The urban theory also suggests that in the initial stages of development, much of the urban population growth will be a natural increase and, to some extent, rural-to-urban migration, which is real urbanisation. Population growth through natural increase is just urban population increase, not urbanisation.
In India, rural-to-urban migration is low compared to other countries because the languages, religions and habits differ. A person from UP or Bihar would feel very uncomfortable migrating to South unless it is because of economic compulsions. Interstate migration in India is among the lowest in the world.
Much of the migration takes place within the state. Rural-to-rural migration is very high because of marriages. Urban-to-urban migration is now picking up from small to large or medium-sized cities. However, the levels of rural-to-urban migration in India have been maintained at about 20-21%.
We haven’t had a census yet, but would the next one show similar trends?
The trend should continue. So far, nothing suggests that it would change. The only pattern that could change is incorporating rural areas into urban areas.
The share of census towns will come down. This is the only part of the demography that we can project from the older data. All the projections, including the Centre for Policy Research’s paper, say that the proportion will decrease. So, all this noise we’re making now will die down unless we change the definition (of urban).
Why are the number of census towns likely to come down?
The rural-urban distinction has become very thin. In Delhi, for instance, Najafgarh, Narela, and even Mehrauli were seen as independent of the core Delhi. Now, they are all integrated. The distinction between rural and urban begins to thin as development advances.
This is a natural process that is happening in some parts, but not in others. If you travel from Chennai to Puducherry or Vijayawada to Vizag in coastal Andhra Pradesh, you will hardly see a village like in Bihar, UP, or Rajasthan.
What are your thoughts on the spatial dynamics of urban growth and in-situ urbanisation?
These areas should be treated as urban rather than rural. You can call them by any name, but integrated development will not be possible unless the census towns are also integrated within the metropolitan area, even if it is not a metropolitan area but a city of 100,000. We are failing on that account. We have not even started.
Why do we get surprised when the rural area changes its character? We must give it what the new character requires. Why are we resisting it?
But we are reluctant to do that for the reasons I explained earlier. We treat rural and urban differently. The spatial angle in our policy is virtually zero. A very conscious attempt by the Rural Development Ministry to see those census towns do not ask for conversion into urban; they brought in a scheme that integrates them into RURBAN clusters so that these places remain within the ambit of the ministry. These anti-urban activities are also costly.
Does India have a standard, common narrative on urbanisation now?
The standard narrative we keep hearing is that India is urbanising at a very rapid rate; we produce 60-70% of the GDP. But we are neither growing at such a fast rate nor is our GDP in cities 60-70%. The papers in my book use formal data sets that are comparable over time.
No matter what, India will move from a lower-middle urban economy to an upper-middle urban economy by 2047. Whether it will be a productive and inclusive transition depends on how we handle it.
Economic growth was 7%—8% in the last quarter. It’s driven in large part by capital, innovations, and technology. However, two of the four factors of production—land and labour—are expanding very weakly to support this growth.
China’s urbanisation is driven by land expropriation. They buy village after village and incorporate them into cities. Housing (sector) is becoming a liability now in China, essentially because they overdid the expropriation and the building activity. Their housing (inventory) is empty (unoccupied). That is one extreme.
In India, it’s the opposite. Land regulation in India is among the stiffest anywhere in the world. Only 3% of the total land is in urban areas, which hold 35% of our population. Land regulations—the zoning, height, and Floor Area Ratio regulations — ought to be changed and were discussed in the second five-year plan (1956-61) and every plan up to the 12th (2012-17).
The total number of people in our labour force is low. The formal part is those (accounted for) in the Annual Survey of Industries (ASI) and those registered under the Shops and Establishment Acts, wherever these exist. The rest is all informal.
The formal part is small because our industrial policy treats an industry with nine people and electricity (connection) or 20 persons without electricity (connection) as unrecognised. As soon as they cross the line, they come under the ASI and must register. So, they’re happy running two, three or four unregistered companies because they don’t have to pay the registration fee. Their contribution is not even counted in the ASI numbers. And our informal, unorganised sectors continue to produce inferior-quality goods.
In State of the Cities report (2021), you pointed out how informality in the labour market in cities is growing while formality is increasing in the rural sector. Can you explain?
They are all moving to the census town because they do not have to pay taxes there. Why would they remain in proper Delhi if they can go to areas treated as rural? Formalisation of the rural economy affects productivity. This is a policy-driven constraint that prevents them from exploiting their potential in a place where they ought to be.