A viral pavement row put one homeless family in the political spotlight. But with no current count of Kolkata’s homeless population and an official shelter capacity of just over 1,000, the larger question remains: what happens to those who are told to leave the street but have nowhere else to go?
Kolkata: Kolkata’s homeless population has been counted, surveyed and legislated for more times than it has been housed. A minister’s reprimand on Free School Street recently reopened a question the city has never satisfactorily answered: where are people living on its pavements supposed to go? The controversy began when West Bengal Urban Development and Municipal Affairs Minister Agnimitra Paul, during an inspection of the Free School Street-Mirza Ghalib Street area, encountered a pavement-dwelling woman heating milk on a kerosene stove. Paul asked her to extinguish the stove and clear the pavement. The woman, identified in several reports as Sita Devi, responded with a question that quickly became the centre of the controversy: why would anyone choose to live on a footpath if they had a home?
The exchange was filmed and circulated widely. It triggered criticism of the minister as well as a political contest over the family’s rehabilitation. Days later, Chief Minister Suvendu Adhikari met Sita Devi and her family at his public grievance programme and directed that alternative accommodation be arranged. The administration subsequently said accommodation had been provided at a location chosen by the family and that Sita Devi had also been offered work as a cleaner at a restaurant. For one family, public attention produced a relatively swift intervention. But the episode raises a more difficult question about the pavement dwellers whose circumstances never become viral videos.
Kolkata’s history makes the issue particularly complex. Large parts of the city and its fringes absorbed successive waves of migrants and refugees, particularly after Partition. Refugee families occupied vacant land, railway embankments and canal-side areas, and many informal settlements were regularised only over subsequent decades. Later governments concentrated substantially on improving bustees through water supply, drainage, roads and other civic facilities. But people living on pavements, under flyovers, near railway stations and outside established settlements remained a different and often less visible population.
The national policy response came through the Scheme of Shelter for Urban Homeless under the Deendayal Antyodaya Yojana-National Urban Livelihoods Mission. Its guidelines envisaged permanent, all-weather shelters open round the clock. For every one lakh of urban population, provision was to be made for community shelter space for at least 100 people, with individual shelters generally accommodating between 50 and 100 persons. DAY-NULM itself ended on September 30, 2024, although shelters created under it continue to operate. The Union Housing and Urban Affairs Ministry told Parliament this year that 1,995 shelters with a combined capacity of about 1.41 lakh beds had been created nationally during the mission period.
Kolkata Municipal Corporation’s own 2026–27 Budget gives a more precise picture of the city’s present infrastructure. Its Social Welfare and Urban Poverty Alleviation Department operates 12 Shelters for Urban Homeless with a combined capacity of 1,035 people. Two more completed shelters, in wards 92 and 95, are expected to accommodate another 126 people, while four shelters under construction are projected to add 447 places. Yet there is no reliable current figure against which that capacity can be measured.
The 2011 Census remains India’s last full national census baseline for homelessness. A subsequent ward-level night survey conducted in 2014 by Sabuj Sangha with the Indo-Global Social Service Society counted 39,431 homeless people across 118 of the then 141 KMC wards. Different studies and surveys have since produced widely varying numbers, reflecting differences in definitions, geography and methods as well as the highly mobile nature of the population. What Kolkata lacks today is a current, comprehensive enumeration capable of answering a basic planning question: how many people actually require shelter or rehabilitation, and where are they?
TheNews21 sent two rounds of questions between August 29 and September 9, 2026, to the West Bengal Department of Urban Development and Municipal Affairs and the Kolkata Municipal Corporation, seeking current figures for the homeless population, shelter occupancy and the administration’s response to concerns over rehabilitation. Neither responded. Officials approached at KMC headquarters also declined to comment, referring to a West Bengal government circular issued on May 19, 2026, restricting unauthorised interactions by government employees with the media and disclosure of official information. The circular does reiterate restrictions on government employees communicating official information to the press without authorisation.
Statistics, however, explain only part of what happens between the street and the shelter. Parul — a pseudonym used at her request — knows that journey from experience. She lived with her children on a pavement near Sealdah station for about six years, from 2014 until the COVID-19 pandemic. Her family had lost a rented room after her husband’s construction work dried up. She has since moved away from the pavement, lives in Basanti Colony and works as a domestic help. Recalling her years on the street, Parul describes the repeated instructions to pavement dwellers to move on. “They come with the loudspeaker and say, clear the road, clear the road. Where do they think we should clear it?” she says.
She also recalls approaching a shelter near the station while she was homeless. According to Parul, her teenage son could not stay with her. “They would not take my son. He was fifteen; they said he was not a child anymore. So we stayed outside.” Her account is an individual experience from those years, rather than evidence of a current universal shelter rule. But it illustrates how family circumstances can determine whether a shelter is actually usable.
Bijoy — also a pseudonym — says he has been living under the Gariahat flyover for nearly 15 years. He recalls an eviction drive two winters ago that, he says, was not followed by an offer of accommodation. “They took my stove, everything went in the truck. No one came back to take us anywhere or offer us another place to live,” he says. “We just came back after a week when the drive was over.”
For Asmita Saha, an NGO worker who has worked with children in Kolkata Municipal Corporation homeless shelters, such experiences demonstrate why homelessness cannot be separated from employment, family and neighbourhood networks. “Homelessness is usually the result of several vulnerabilities coming together,” she says. Families depending on domestic work, construction, street vending, driving or other daily-wage employment can lose the ability to pay rent after even a relatively short interruption in earnings. Moving into a shelter is therefore not necessarily a straightforward transition. “One major difficulty is trust,” Saha says. People who have spent long periods in informal settlements or on the streets may be uncertain about a shelter’s rules, privacy, length of stay or treatment. Parents may also worry about whether they can continue working and whether their children can remain in the same schools.
Location is crucial. “For some people, the street is actually closer to their livelihood,” she says. Moving a street vendor, construction worker, rickshaw puller or casual labourer far from the area where work is available can increase transport costs and reduce earnings. “A shelter can be structurally adequate but still fail a family if it is disconnected from the ecosystem that supports their lives,” Saha says. If moving into shelter means a parent loses access to employment or a child loses access to school, the intervention may create another form of instability. That also changes the way success should be measured. Saha argues that the number of buildings or beds tells only part of the story. The longer-term questions are how many people move from shelters into stable housing, how many find sustainable employment, how many children remain in school and how many families eventually become independent of institutional support. “The goal should not be simply ‘how many homeless people can we shelter’ but rather ‘how can we help families move from homelessness to stability and independence’,” she says. “That shift from shelter as an endpoint to shelter as a transition phase is what would make the system genuinely sustainable.”
Architect and urban-planning professional Rishav Ranjan Roy makes a similar argument from the perspective of city design. He cautions against creating a large rehabilitation colony at the urban periphery and moving pavement dwellers there simply because land is cheaper. Kolkata’s pavement-dwelling population, he says, has historically clustered around areas of economic activity. Daily-wage workers, hawkers and waste pickers depend on proximity to markets, transport nodes and central commercial districts. Roy favours a distributed network of smaller shelters and affordable-housing options close to employment and public transport rather than concentrating rehabilitation in distant locations. “Rehabilitation shouldn’t just mean a bed and a roof,” he says. It should also account for sanitation, cooking facilities, healthcare, childcare where necessary, storage, documentation and connections to livelihood programmes.
He also distinguishes between making public space more inclusive and treating public space as a substitute for housing. Toilets, drinking water, shaded areas, seating and safer night-time environments can make a city more humane, but they cannot resolve homelessness. “You cannot solve habitation purely through public-space design,” Roy says. “People need actual housing, and public space should complement that housing.” Roy believes elements of the “Housing First” approach used internationally are worth examining, although he cautions against importing a Western model wholesale. In a dense city such as Kolkata, he suggests that existing housing, smaller apartment buildings, adaptive reuse and infill development may be more practical than vast rehabilitation estates.
The central question remains location. A house may have little rehabilitative value if reaching work requires additional hours of travel and unaffordable daily transport costs. Kolkata, then, is not without a shelter policy. Its municipal budget shows an existing network and plans for expansion. The problem is that the city still lacks a current count of the people that network is supposed to serve, while field accounts show that access to a bed cannot by itself measure whether rehabilitation has worked. The debate is sometimes presented as a choice between keeping pavements clear and showing compassion towards people who live on them. That is a false choice. Footpaths are public spaces and pedestrians have a legitimate right to safe access. But removing someone from a pavement does not answer the next question: where can that person go without losing family, work, schooling or the possibility of earning a living?
Sita Devi’s family came under intense public attention and, according to the government, received accommodation and employment. Parul eventually left the pavement and has rebuilt a working life. Bijoy says he remains under a flyover. Their circumstances are different, but together they point to the same gap between clearance and rehabilitation. The question for Kolkata is no longer simply whether people should live on its footpaths. It is whether the city can offer a credible, dignified alternative before asking them to leave.
