Number crunchers and slum sums
Differences have broken out within the government on how many slum dwellers India really has
A disagreement has erupted between the urban development ministry and the census commissioner’s office over the recently released provisional results on slum population in India. At stake are the huge amounts doled out for slum development in the country.
For the first time, the census has attempted to collect data on the slum population of the country and put the number of slum dwellers at 40.6 million.
The urban development ministry has, however, chosen to reject the provisional data and termed it “defective” and “faulty”. The ministry claims that the slum population in the country should be between 61 million and 70 million in 2001.
The figure is important because it is used as the basis for preparing slum improvement and urban poverty alleviation programmes. In this case, the issue was raised when the fund allocations for the Valmiki Ambedkar Malin Basti Awas Yojana for urban slum dwellers, a scheme worth roughly Rs 3,000 crore, was under discussion.
At the crux of the dispute over the numbers are two issues — the definition of a “slum” and the methodologies adopted to estimate slum population.
For purposes of the census, the census commissioner has collected information on slums only in cities that reported a population of over 50,000 in the 1991 census. These comprise what are referred to as Class I and Class II cities.
Areas considered slums are those that are notified as a “slum” by state or local governments in cases where a Slum Act exists. Besides this, areas recognised as a slum by state or local governments, but which may not have been formally notified as a slum under any Act, are also included.
The rule of thumb followed in the case of areas where no formal notifications were available was that at least 300 out of the 500 people in the enumeration block should be living in poorly built, congested tenements, in an unhygienic environment, with inadequate infrastructure and lacking in proper sanitary and drinking water facilities. On this basis, the census for 2001 puts the number of slum dwellers in 743 towns in the country at 40.6 million.
The urban development ministry, however, claims that the census count is way below the 1997 National Sample Survey estimates of urban poor. The estimates put the number of urban poor at about 70 million, which is 24.6 per cent of the total urban population of 285 million. Since most of the urban poor live in urban slums, the census figure of 40.6 million is a gross underestimate, the ministry has pointed out.
Also, since the census figures state that there is no slum population in Lucknow, the ministry has said that the data is unreliable. It has, in fact, called for another survey to get a proper figure of the percentage of population living in slums.
In view of the “faulty” population figures of the census of 2001, it has also suggested that the next best alternative would be to use the projected figure of slum population for 2001 prepared by the Town and Country Planning Organisation (TCPO).
The TCPO had estimated that the slum population in India in 2001 would be 61.8 million, or about 21.3 per cent of the total urban population of 291 million.
This is a point of view, that has been strongly disputed by the census commissioner. For one, he points out that it is simplistic to assume that most of the urban poor live in slums or that slum dwellers in urban areas are poor. In cities like Delhi and Mumbai, most of the slum dwellers are likely to have income levels that put them way above the poverty line.
Moreover, the figures put out in the provisional results of the census relate to only 607 large towns and a comparison with the estimated 7 crore total urban population of the country is not proper.
J K Banthia, the registrar general and census commissioner of India, says that the census in urban areas was carried out through the municipal administration or urban authorities, the list of notified and recognised slums was obtained from them and the census officials were also appointed from the same agencies.
Since local perception is important when identifying a slum, the census authorities relied on local authorities to do so. “Out of a total of 743 cities that had a population of over 50,000 in 1991, only 607 reported slums. This means that town authorities in as many as 136 towns chose not to identify any area in their jurisdiction as a slum,” he says.
Lucknow was one example where no slum population was reported by the Mahanagar Adhikari, who is the highest executive authority of the municipal corporation.
“Since such a situation seems unlikely, we have asked them to review the situation and get back to us. Even with the provisional figures, we have included a caveat to that effect,” Banthia says, adding, “However, our job is only to count the people living in areas that the local authorities identify as slums.”
Banthia says reviews have also been requested in the other towns that reported no slums. “If they report any slum areas, we will go and conduct a head count. The changes will be included in the final figures we compile,” he says.
Also, he points out, in the absence of a slum act for each state, there is bound to be an element of subjectivity in perception as to what constitutes a slum. Perceptions may vary from state to state, and even from town to town. What constitutes a slum in Mumbai could well be a “well developed” area in some other parts of India.
Most states in the north east do not have a Slum Act. Also, very few towns there would have a population of over 50,000 people. As a result, Sikkim, Arunachal Pradesh, Nagaland, Manipur and Mizoram have reported no slum population.
On the TCPO estimates, Banthia points out that they are based on ad hoc calculations based on various surveys at different points of time. The estimates are based on the slum population figures supplied by the states and Union Territories, all of whom follow different estimates for the base figure.
“A Compendium on Indian Slums”, brought out by the ministry of urban affairs and employment in 1997 has specified the methodology.
In the case of Assam, Andaman & Nicobar Islands, Delhi, Himachal Pradesh, Rajas-than, Meghalaya, Pondicherry and Sikkim, which had furnished the figures of total estimated/identified slum population for 1991, the projections for 2001 were based on the 1991 slum population percentage.
For Andhra Pradesh, Gujarat, Haryana, Orissa, Punjab, Tamil Nadu, Tripura and Uttar Pradesh, the projections were based on information on their Class I and Class II urban settlements in 1991.
For Kerala and Madhya Pradesh, it was on the basis of the 1981 figures, while those for Karnataka were based on the report of the Task Force on Housing and Urban Development, Shelter for the Urban Poor (1983).
The problem with basing projection on such old data is self evident. An area notified as a slum in 1981 and subsequently provided with the requisite amenities may no longer be a slum. Sections of the walled city area of Delhi, for instance, are still notified as slums whereas they are fairly well developed now. There is no process of denotification of slums, says Banthia.
Also, especially in Delhi, the government has gone on a number of demolition drives. Even if the slum is technically not present, the slum data with authorities is not updated, resulting in an overestimation of the slum figures with which state governments are comparing the census figures, he adds.
Despite this, says Banthia, the TCPO estimates of the proportion of slum population living in Class I and Class II towns is 22 per cent, which is lower than the count provided by the Census. “The census is a head count, and by far the most reliable source of data available at any point in time,” Banthia insists.
Officials dealing with urban development are, however, unconvinced. Even if the census figures are more reliable than other estimates available with the government, and they accept the argument that the data deals with only 743 cities larger cities in India, the census still does not give a complete picture of the slum situation, they feel.
The definition of a slum adopted by the census authorities requires at least 300 people (60 per cent) in an enumeration block to be living in slum-like conditions for the block to be counted as a slum. This leaves out large sections of the population living in these towns who are living in slum-like conditions.
“If 250 or even 25 people are living in a jhuggi cluster in a city, their living conditions need to be improved and the government still has to identify them and provide for it,” they say.
Also, slums need not be present only in larger towns. Small towns also
have slum populations and they have to be identified too. The census figures
are useful only to a limited extent to them, they say.