Policy Advocacy for Pavement Dwellers in Mumbai

In India’s biggest cities, the most visible form of poverty is often the one people try hardest to overlook. Families cook, sleep, study, and raise children on narrow strips of footpath wedged between traffic and shopfronts. For decades, pavement dwellers in Mumbai were treated as an eyesore rather than as residents with rights. That perception began to shift in the 1980s, when a small group of women, activists, and researchers started doing something deceptively simple: counting people, listening to them, and putting their lives on the public record.

Who pavement dwellers actually are 🔗

To understand the policy advocacy that followed, it helps to first understand who these families are. Pavement dwellers are mostly first-generation migrants who arrived in Mumbai as early as the 1940s and have lived on the pavements of public roads ever since. Their shelters are often made of cloth, cardboard, corrugated iron, or scraps of wood, built against compound walls that separate private plots from the street. Many belongings are kept in nearby shops or kiosks because the pavement itself offers no safe storage.

They are not drifters. Roughly half came from the poorest districts of Maharashtra, and the rest from the poorest parts of the country, pushed by drought, famine, earthquakes, riots, or the collapse of rural livelihoods. Yet for much of the twentieth century, they were effectively invisible to the state: no census counted them, no welfare scheme reached them, and no housing policy imagined them as beneficiaries.

The 1985 turning point: Olga Tellis and the threat of mass eviction 🔗

In 1981, the Maharashtra government and the Bombay Municipal Corporation announced a sweeping eviction drive. The state decided that pavement and slum dwellers should be cleared from the city and “deported to their respective places of origin”, using powers under the Bombay Municipal Corporation Act of 1888. A writ petition filed by journalist Olga Tellis on behalf of the dwellers reached the Supreme Court, and the judgment came down in 1985.

The ruling was mixed but historic. The Court held that the right to life under Article 21 of the Constitution includes the right to livelihood and shelter, and insisted that natural justice and a fair hearing had to precede any eviction. At the same time, the Court granted the Municipal Corporation of Greater Mumbai the authority to demolish household structures on the city’s sidewalks.

In other words, eviction could happen, but only with due process. Whether that process would actually protect the poorest families depended on who could speak for them. That vacuum is what civil society stepped into.

Enter SPARC and the Alliance 🔗

The Society for the Promotion of Area Resource Centres, known as SPARC, was founded in 1984 by Sheela Patel and a small team working out of Byculla in central Mumbai. SPARC chose to begin with the most vulnerable and invisible group of the city’s urban poor, the pavement dwellers, on the belief that solutions designed for the poorest could later be scaled up for others. Within two years, SPARC had partnered with two grassroots formations: the National Slum Dwellers Federation (NSDF), a pre-existing federation of community leaders from informal settlements, and Mahila Milan, meaning “Women Together”, a network of women’s collectives from pavement communities.

Together, these three organisations became known simply as the Alliance, and over thirty years they worked with thousands of households living under the constant threat of eviction, helping them secure permanent housing and basic services. Mahila Milan is especially significant: it was formed in 1986 when around 500 women living on Mumbai’s pavements organised themselves to prevent the demolitions of their homes, shifting leadership in these communities firmly into women’s hands.

Why women-led collectives mattered 🔗

Many groups responding to the 1985 demolitions initially planned confrontational protests. But SPARC found that women from pavement settlements did not want confrontation; they preferred to find a way for the city to coexist with them. That insight reshaped the advocacy strategy. Instead of street battles, the Alliance invested in surveys, savings groups, dialogues with officials, and model housing exhibitions – slower work, but work that built durable relationships with the state.

Research as advocacy: “We, the invisible” 🔗

The single most influential tool the Alliance developed was a community-led census. In 1985, SPARC partnered with the Society for Participatory Research in India to carry out a census of around 6,000 pavement-dwelling families, published as the report “We, the Invisible”. The census was conducted transparently, with the households themselves kept informed at every stage, and pavement dwellers trained to collect the data about their own neighbourhoods.

The numbers transformed the debate. The exercise covered over 6,000 households and nearly 27,000 people, enabling pavement dwellers to become formally recognised and visible to authorities and to the rest of the city, and setting an advocacy base for policy change. For the first time, officials could no longer dismiss pavement families as a faceless nuisance; here were names, occupations, years of residence, places of origin, and household sizes.

The report also challenged a popular myth. It revealed that pavement dwellers were not transient populations but people who had lived for over two decades in the city. This finding directly undercut the argument that they were recent “encroachers” who could simply be bussed back to their villages.

Why the census worked as advocacy 🔗

The census did three things at once. Because it employed formal data collection methods and a reporting style that policymakers trust, it became a means to communicate with government at various levels. It also built organisational capacity inside the community, training residents to negotiate with officials using their own data. And it created a shared identity across pavement settlements that had previously been organised only along lines of kinship, language, or region of origin.

Community-led enumeration has since become a signature strategy of the Alliance, and has been taken up by slum federations across India and internationally.

From counting to co-designing: housing exhibitions and model homes 🔗

Visibility alone does not deliver housing. The Alliance’s next innovation was to flip the usual process of slum rehabilitation on its head. Instead of letting engineers and officials decide what poor families “needed”, they asked the families themselves to design their homes.

After an early resettlement in Goregaon left families unhappy with decisions made on their behalf, SPARC built four model houses and displayed them in Byculla for a week, inviting everyone who had participated in the census; over 5,000 households attended, and the most popular model was the most expensive one, which was then taken to more than fifty similar exhibitions. These housing fairs turned abstract debates about “affordable housing” into concrete choices that residents could walk through, touch, and argue about.

Pushing the policy needle: from demolitions to dialogue 🔗

Sustained advocacy eventually reshaped the formal legal framework for housing the urban poor in Maharashtra.

The Slum Rehabilitation Act, 1995 🔗

After a decade of surveys, negotiations, and model projects, the Maharashtra government overhauled its approach. The Slum Rehabilitation Act of 1995 was the product of policy development that included grassroots slum dweller organisations, particularly SPARC, and through the Act pavement dwellers were for the first time accepted into the category of households entitled to land for relocation. Anyone who could prove residence in Mumbai before 1 January 1995 was protected from eviction, and a Slum Rehabilitation Authority was set up to oversee the rebuilding process.

Around the same time, the scheme gave each eligible family the right to a self-contained rehabilitation tenement of 225 square feet of carpet area, free of cost, with developers being granted land-use incentives for implementing projects. The structure wasn’t perfect, but for the first time the state had formally acknowledged that pavement families had a claim on the city.

Relocation and the Mumbai Urban Transport Project 🔗

The clearest demonstration of what Alliance-led rehabilitation could look like at scale came with the Mumbai Urban Transport Project. In total, 20,000 people were relocated from the railway edges under the project, and 17,000 of them were supported through the relocation and rehabilitation process by SPARC, NSDF and Mahila Milan. The federations’ years of survey work turned out to be decisive: they already knew which land parcels were available, who lived where, and what households needed.

This experience also reshaped international policy. When the Indian government initially fell short on World Bank requirements around shelter-before-demolition, the once-temporary Mankhurd resettlement site became a permanent one, providing lasting housing for displaced rail-side families.

Sanitation as a parallel front 🔗

Housing was never the only demand. The Alliance also pushed hard on sanitation, designing and constructing community toilet blocks in cities across the country in partnership with municipalities. This work showed that community organisations could not only advocate for services but actually deliver them at scale – a powerful argument for including civil society in the design of urban policy rather than merely consulting it at the end.

What this case teaches about civil society and policy-making 🔗

The pavement dwellers’ story is often used in public administration classrooms, and for good reason. It shows how civil society organisations can shape policy without elections, courts, or street agitation being the only tools available.

A few patterns stand out. First, evidence changes conversations. The 1985 census reframed pavement dwellers from “encroachers” into long-settled residents with verifiable histories. Second, organised communities are more persuasive than individual petitioners. Mahila Milan and NSDF gave officials a stable counterpart to negotiate with, which made compromise possible. Third, co-production beats consultation. By designing model houses, surveying land, and later building toilets and tenements themselves, the Alliance blurred the line between advocate and implementer. Fourth, incremental wins compound. Each round of negotiation – from the Olga Tellis hearing to the 1995 Act to the MUTP rehabilitation – widened the legal and administrative space for the next one.

None of this means the problem is solved. Evictions without proper resettlement continue in many Indian cities, and the Slum Rehabilitation Scheme has been widely critiqued for slow implementation and uneven quality. But the shift from invisibility to recognition, from bulldozer to negotiation, is a real one, and it was driven as much by women in Byculla with survey forms as by lawyers in Delhi with writ petitions.

What do you think? If a policy decision is going to affect thousands of families, who should produce the data that informs it – the government, independent researchers, or the affected communities themselves? And what does this case suggest about the limits of court judgments when they are not backed by organised community advocacy on the ground?

References

1. https://en.wikipedia.org/wiki/Pavement_dwellers
2. https://www.drishtiias.com/daily-news-analysis/olga-tellis-case-1985
3. https://legalfly.in/olga-tellis-v-bombay-municipal-corporation/
4. https://en.wikipedia.org/wiki/Olga_Tellis_v._Bombay_Municipal_Corporation
5. https://www.sparcindia.org/aboutsparc.php
6. https://sparcindia.org/about.php
7. https://en-academic.com/dic.nsf/enwiki/5606329
8. https://en.wikipedia.org/wiki/We,_the_Invisible
9. http://www.communityplanning.net/special/makingplanningwork/mpwcasestudies/mpwCS05.php
10. https://www.primidi.com/pavement_dwellers
11. https://sparcindia.org/pdf/articles/ennumerations.pdf
12. https://en.wikipedia.org/wiki/Slum_Rehabilitation_Act_1995
13. https://assets.publishing.service.gov.uk/media/57a08d7de5274a27b2001889/R7394paperHI_SPARC1_FINAL__Jan00.pdf
14. https://sdinet.org/tag/sparc/