English only · Odia translation in progress

Chapter 3: The Other Odisha in Surat

On May 1, 2020, a video circulated on Odia news channels that condensed an entire migration story into thirty seconds. Thousands of workers — mostly young men, some carrying children, all carrying bags on their heads — crowded the platform at Udhna Junction in Surat, pushing toward a Shramik Special train bound for Odisha. The platform announcement, unusually, was in Odia — because Udhna serves a neighborhood where Odia is the lingua franca. The men wore the exhaustion of people who had not worked in five weeks and could not afford another day of not working. Some had walked from powerloom dormitories in Ved Road and Limbayat. Some had pooled money for autorickshaws. All of them were trying to get home to Ganjam, a district in southern Odisha that most Indians could not locate on a map but that had quietly supplied the labor force for India’s largest polyester production center for the better part of a century.

By July 7, 2020, the government of Odisha had received 3,58,401 people back by special trains alone. Of those, 1,30,537 — thirty-six percent, the single largest share from any state — came from Gujarat. The total number who returned by all modes of transport by mid-June was 5,43,905. And that was a fraction. When the state government opened a portal for stranded migrants to register, ten lakh people signed up — a million people, many without smartphones, many without internet access, many without awareness that the portal existed. The actual number of Odias in Gujarat was, and remains, unknowable with precision. But the COVID-19 exodus did what no census had managed: it made visible a population that India’s largest democracy had been content to not count.

The number most commonly cited is seven to eight lakh. Between 500,000 and 800,000 Odias live and work in Surat, the overwhelming majority from a single district — Ganjam. If Surat’s Odia population were a city, it would be roughly the size of Bhubaneswar. It would be the largest concentration of Odias anywhere outside Odisha. It would have its own schools, its own festivals, its own social hierarchies faithfully transplanted from villages 1,600 kilometers away. It would have more than five hundred community organizations. And it would have zero political representation — no elected voice in Gujarat, and an electoral weight in Ganjam that evaporates every time a voter is eight hundred miles away on election day.

This chapter is about that city within a city. How it came to exist, what it produces, what it costs the people who sustain it, and why, after eighty years, it shows no signs of being a temporary arrangement.

The Eighty-Year Pipeline

Migration corridors do not appear suddenly. They are built, person by person, over decades, until the path between origin and destination becomes as fixed as a road — invisible on any map, but more reliable than most infrastructure the government has ever built.

The Ganjam-Surat corridor has its roots in the 1940s, possibly earlier. The initial migrants were not powerloom workers. They were landless Dalit laborers, primarily from the Kewat caste — traditionally fishermen and boatmen — who first migrated to rice mills in Myanmar and jute mills in West Bengal. This was colonial-era labor migration, driven by the same forces that moved millions of Indians across the subcontinent and beyond: landlessness, poverty, and the existence of someone who had already gone and could vouch for the next arrival.

By the 1970s, the destination shifted. Surat’s textile industry was expanding, and workers from Ganjam began arriving for gardening and construction work — not yet the powerloom labor that would later define the corridor, but the first thread in what would become an industrial-scale supply chain of human beings.

The critical turning point came in the 1980s. Two transformations converged. First, the powerloom industry began migrating out of Mumbai — rising rents, labor regulations, and the mill closures of the 1980s pushed production to Surat, where land was cheaper, regulation lighter, and electricity more reliable. Second, agriculture in Ganjam was hit by a succession of natural disasters — cyclones, floods, droughts — that made farming a losing proposition for families operating on marginal land. As IndiaSpend documented, “As these transformations unfolded both at the source and the destination, the two trends converged, prompting people from Ganjam to start migrating to Surat.”

The mechanism was chain migration, and it worked exactly the way chain migration always works. One man goes. He finds a job through a contact — a fellow villager, a caste member, someone who can vouch for him to a loom owner. He sends word home. His brother follows. His cousin follows. The cousin brings a neighbor. The neighbor brings his wife’s brother. Within a generation, entire villages have a presence in Surat. Within two generations, the corridor is self-sustaining. A young man in Surada block doesn’t need a government employment scheme or a recruitment agency. He needs a phone number — and the number belongs to someone from his village who already knows which loom owner needs workers and which dormitory has an open bed.

This is what makes the corridor indestructible by policy. It is not a government program. It is not an employer initiative. It is a social institution, built from trust networks that operate along the oldest organizational lines in Indian society: caste and kinship.

By the 1990s, the composition of the corridor shifted. OBC and general caste communities from Ganjam — Teli, Gouda, Khandayat families who had been landholders but faced agricultural distress — joined the migration stream. The corridor became OBC-dominated. And as it did, the Dalit communities who had been early migrants found themselves marginalized — not by Gujaratis, but by their own caste hierarchy, reproduced 1,600 kilometers from home.

The Numbers

No one knows exactly how many Odias live in Surat. This is itself a fact worth pausing on. A city hosting somewhere between half a million and eight hundred thousand people from a single Indian state does not have a reliable count of those people. They are not registered as interstate migrant workers under any law. They do not appear in Gujarat’s voter rolls. Many are not registered under their real names at their workplaces. They exist in a statistical shadow — present enough to operate ninety percent of India’s polyester production, absent enough to elude every counting mechanism the state has devised.

The commonly cited figure of seven to eight lakh comes from a convergence of estimates: a 2007 UNDP study, subsequent NGO assessments, and the COVID-19 return data that revealed the lower bound of the real number. The Gram Vikas-CMID block migration profiles offer granular evidence. In Jagannathprasad block of Ganjam, a 2020-21 survey of 421 households found that 57.2 percent of people had stayed for work outside the district for thirty or more days in the past ten years. Nearly one-third of those migrant workers went to Surat. In Surada block, a 2023 survey found the same pattern — nearly one-third in Surat — with a staggering additional finding: 95.8 percent of migrant households reported that they could not have escaped poverty without migration income. The highest rate of any block surveyed.

The specific blocks that feed the corridor are identifiable: Surada, Jagannathprasad, Polsara, Chikiti, Aska, Bhanjanagar, Digapahandi, Hinjilicut, Khallikote, Buguda. Ganjam has twenty-two blocks. The Surat pipeline draws from most of them, but the intensity varies. The blocks closest to the coast, where cyclone damage to agriculture is most frequent, send the most workers.

Why Ganjam specifically? The answer is a particular combination of factors that no single variable explains.

First, location. Ganjam is coastal, which means cyclone exposure. The 1999 super cyclone devastated the district. Cyclone Phailin in 2013, Cyclone Fani in 2019 — each one destroyed agricultural livelihoods that took years to rebuild, pushing another cohort of families toward the already-worn path to Surat.

Second, literacy. Ganjam is not among Odisha’s most deprived districts. Its literacy rate is higher than the KBK interior. Its people are educated enough to navigate a city, operate a loom, manage cash, send remittances through banks — but not educated enough, in most cases, to access white-collar work. They occupy the band of Indian society that is too skilled for agricultural subsistence and too unskilled for the formal economy. The powerloom is perfectly calibrated to this band.

Third, traditional textile knowledge. Ganjam has a history of weaving — not powerlooms, but handlooms. The Teli community, which dominates the corridor, traditionally processed oilseeds and had trade networks. The transition from agricultural and artisanal work to industrial textile labor was not as large a jump as it would have been from, say, tribal forest communities.

Fourth, and most importantly: the network. Once the corridor established itself, it became self-reinforcing. A young man from Ganjam doesn’t migrate to Surat because he did a cost-benefit analysis of interstate wage differentials. He migrates because his uncle is already there, because his neighbor’s son went last year and came back with enough money to build a pucca house, because the path is so well-worn that not taking it feels like the riskier choice.

The Powerloom Economy

To understand what seven lakh Odias do in Surat, you need to understand the industry they serve.

Surat produces approximately ninety percent of India’s polyester fabric. The city has an estimated 1.5 million loom machines, spread across thousands of small-scale units. The industry is Gujarat’s largest employer of migrant labor. It is also, by almost any measure, India’s most successful example of a labor-intensive manufacturing cluster — if you measure success by output and ignore what it costs the people who produce it.

The structure is simple and stratified. The powerloom units are owned by Gujarati entrepreneurs, many from the Patidar and other business communities. Units are small — typically four to fifty looms per facility — and largely unregistered. This fragmentation is not accidental. It is the industry’s regulatory armor. Most units are registered under the Shops and Establishment Act rather than the Factories Act, which means they are exempt from factory-grade labor protections: no accident compensation, no insurance, no provident fund, no gratuity, no bonus.

The workers are Odia. Eighty percent of Odia migrants in Surat are from Ganjam. They operate the looms — eight to sixteen machines per worker, depending on the fabric and the unit. The work is piece-rate: Rs 1.10 to Rs 1.50 per meter of fabric produced. A worker running looms for twelve hours produces enough to earn Rs 20,000 to Rs 25,000 per month. This is roughly three to four times what MGNREGA provides in Odisha (Rs 237 per day for twenty-five days equals Rs 5,925 per month). The wage differential is the gravity that holds the corridor together.

There is a further stratification that speaks to the industry’s internal hierarchy. Odia workers from Ganjam operate the older “khat khat” machines — the name is onomatopoeic, after the sound the looms make — producing lower-quality fabric for domestic markets. Higher-grade imported machines, producing for export, are operated by local Gujarati workers. The division maps neatly onto a value chain: Odia labor produces the commodity fabric; Gujarati labor produces the premium product. The wage differential between the two tiers is significant. The architectural logic is familiar from every extractive economy: the labor that is most replaceable captures the least value.

Shifts run twelve hours. Seven a.m. to seven p.m., or seven p.m. to seven a.m. There is no standard day off — workers may get a maximum of two days per month, usually during power cuts when the looms cannot run anyway. No weekly rest, no paid leave. Workers are classified as casual labor even after ten or fifteen years at the same unit. They are not registered under their real names. They carry no ID cards issued by their employers. If a worker is injured on the job, the legal apparatus that might protect a registered factory worker does not apply. He is, in the eyes of the law, not quite there.

Think of this arrangement as a version of what software engineers call a “zero-knowledge proof” — a system designed so that the employer can benefit from the worker’s labor without formally knowing or acknowledging the worker’s existence. The worker runs the loom, produces the fabric, earns a piece rate, and occupies a dormitory bed. But on paper, no formal employment relationship exists. The worker is a ghost in the machine, and the machine produces ninety percent of India’s polyester.

Rooms by the Looms

The phrase comes from a Citizen Matters photo essay, and it is precise. Odia workers in Surat live where they work — not in separate residential areas with a commute, but in rooms attached to or adjacent to the powerloom units. The geography of Odia Surat is the geography of the textile industry: Ved Road, Panch Manzila, Udhna, Limbayat, Pandesara. These are not neighborhoods in the residential sense. They are industrial zones where human beings sleep.

The numbers describe the conditions more efficiently than adjectives can. A room is typically 500 to 800 square feet. It accommodates sixty to one hundred workers across two shifts — while one shift works, the other sleeps. Per-person space: six by three feet, shared with a shift partner. The buildings are the infamous “Panch Manzila” — five-story structures, dilapidated inside, with narrow corridors whose walls have accumulated years of grime until they are black. Ventilation is poor. Many rooms are windowless. In summer, temperatures reach forty degrees Celsius with inadequate fans. Water supply is intermittent. Two toilets are shared per room. The smell of urine is pervasive — reporters consistently note it as the first sensory impression.

Workers eat at community messes run by fellow Odia migrants. The mess system is a micro-economy unto itself: Rs 2,300 to Rs 3,300 per month covers rent and meals. The price you pay depends on your caste. But more on that shortly.

Scroll.in’s investigative team described the Panch Manzila building on Ved Road: “The dark and narrow corridor of Panch Manzila has years of dirt accumulated on its now black walls.” They documented beds with visible bug blood stains on the walls — from termites, rats, and bed bugs. Workers cannot stretch their arms in cramped 1,000-square-foot units housing eighty to a hundred workers per shift.

What these conditions produce, besides fabric, is disease. Tuberculosis is common. Scabies, fungal infections, malaria. Workers’ diets at the messes are nutritionally catastrophic: a study found daily fat consumption at 294 percent of recommended levels and salt at 376 percent. Hypertension and poor lipid profiles are widespread. The constant exposure to fiber dust from the looms causes respiratory disease. An audiometry study by PARI found that 95 percent of tested workers showed varying levels of deafness from continuous exposure to noise exceeding 110 decibels.

Between 2012 and 2015, just in Surat’s registered powerloom units — the ones that bother to report — 84 fatal events killed 114 workers and seriously injured 375 others. One loom master from Ganjam reported conducting twenty-seven funerals in a single year. Documented deaths include workers as young as sixteen (electrocution) and eighteen (fever and dysentery). The Odisha Labour Minister reported to the state assembly that 403 migrant workers from Odisha had died in other states over the past decade, with Ganjam recording the highest number at fifty-nine.

These are not statistics from a country at war. They are the operating costs of a billion-dollar textile industry, paid in the bodies of people from a single Indian district.

Caste as Architecture

If you want to understand the Surat corridor, you must understand that it is not a labor market. It is a caste institution.

This is the finding that recent investigative journalism — particularly IndiaSpend’s data journalism and Scroll.in’s multi-part investigation in 2024 — has documented most thoroughly, and it overturns the simpler narrative of economic migration that most policy discussions assume.

Nearly seventy percent of Ganjam’s population is OBC and general caste. Scheduled Castes are roughly twenty percent; Scheduled Tribes about three percent. The Surat corridor’s composition mirrors this hierarchy with precision. OBC communities — primarily Teli, Gouda, and Khandayat castes — dominate the pipeline. They control the networks through which jobs are allocated. Getting a position at a powerloom is not a matter of showing up and asking; it is a matter of being vouched for by someone already inside the system, and that someone is almost always a caste member.

The mechanism operates through the mess system. Mess operators recruit workers from their own caste networks in Ganjam. A Teli mess operator brings Teli workers. A Gouda mess operator brings Gouda workers. The mess is simultaneously housing, canteen, and labor exchange. When a loom owner needs workers, he contacts a mess operator, who supplies them from his network. The entire hiring infrastructure bypasses any formal channel — no job advertisement, no application, no interview. It is kinship-mediated labor allocation, as old as India itself, operating inside a twenty-first-century industrial city.

For OBC and upper-caste workers from Ganjam, this system works. It provides a guaranteed job on arrival, a place to sleep, three meals a day, and a social environment of familiar faces. It is, in its way, a support network — a brutal one, but functional.

For Dalits, the system is something else entirely.

Mess operators from upper castes refuse to accommodate Dalit workers. “I ask them their caste when they come,” said Akul Dandapani Nahak, a mess operator who told IndiaSpend he had not accommodated a Dalit worker in twenty years. Food handling is exclusively upper-caste; Dalit cooks are prohibited. When Dalit workers do find accommodation, they get smaller rooms, less space, and worse conditions despite paying comparable rent. SC workers at one documented dormitory shared a room designed for two among six workers — smaller than the hall where upper-caste workers slept.

At work, the discrimination continues. SC migrants face restrictions on which machines they can operate. They are pushed toward the lowest-paying work. The remittance data reflects this: Dalit workers send home an average of Rs 4,814 per month compared to Rs 5,531 for OBC workers. The gap is not because Dalits work less hard. It is because the caste architecture of the corridor assigns them less valuable positions.

The pattern could not be clearer: the hierarchy of Ganjam village is transported 1,600 kilometers and reconstructed in the mess halls, sleeping quarters, and loom assignments of Surat. Caste does not dissolve in the destination city. It rebuilds itself, brick by brick, meal by meal, bed by bed.

The consequence of this is a secondary migration that nobody planned. Marginalized by their own community in Surat, Dalits and tribals from Ganjam have increasingly begun migrating to Kerala instead. Kerala offers two things Surat does not: higher wages (Kerala has among the highest minimum wages in India) and a labor environment where caste discrimination among migrant workers is reportedly lower. Kerala’s own population has migrated to the Gulf in large numbers, creating a domestic labor shortage that Odia workers are filling. The Dalit worker who cannot eat at an OBC mess in Surat can find less hostile conditions in Ernakulam or Thiruvananthapuram.

This secondary migration is the corridor producing its own antibody. The caste system that structures the Surat pipeline is, through its own exclusion, generating an alternative pipeline. Whether the Kerala corridor grows large enough to rival Surat’s remains to be seen. But its existence tells you something about the durability of caste as an organizing principle: it shapes not just who migrates and where, but who is expelled from one migration corridor into another.

Cultural Life in Exile

Despite the conditions — or perhaps because of them — Surat’s Odia population has built a parallel cultural world.

Over five hundred Odia community organizations operate in the city. The sheer number is striking: five hundred groups for a population of seven to eight lakh. It reflects both the community’s cohesion and its fragmentation along village, caste, and neighborhood lines. Each group is a node in a network that connects Surat back to a specific place in Ganjam — this organization draws from Surada, that one from Chikiti, another from Aska. They are not pan-Odia institutions. They are village societies transplanted to an industrial city.

Workers celebrate Nuakhai, Durga Puja, Rath Yatra, and other Odia festivals in Surat. Festival celebrations are often the only time workers get days off from the looms. For powerloom workers who cannot afford the time or fare to return home for festivals, the Surat celebrations are the closest thing to home — a few hours of familiar music, food, and faces amid the twelve-hour shifts and the six-by-three sleeping spaces.

There are Odia-medium schools. Beena Odia Primary School, established in 2005, and Divyajyoti Vidhyalaya serve the community’s children. That Odia-medium primary schools exist in a Gujarati city is itself evidence of the corridor’s permanence — you do not build schools for a temporary workforce. These schools represent a community’s bet that the next generation will also be in Surat, and that it wants its children to know the language of a home they may visit only once a year.

And yet. A notable absence. Despite seven to eight lakh Odia workers, Surat has no formal Odia cultural or community organization of the type that exists in Bangalore (where the Orissa Cultural Association has been active for decades, where Juhar Parivar Bangalore draws ten thousand to Nuakhai celebrations), or Hyderabad (Utkal Parishad), or Delhi (Odia Association). The five hundred groups are informal, village-based, caste-mediated. There is no umbrella organization advocating for the community’s collective interests — no analog to the Kerala Muslim Cultural Centre that represents Malayali workers in the Gulf, no Tamil Sangam that organizes Tamil workers in other states.

This absence is not accidental. It reflects the nature of the corridor itself. The Odia IT professional in Bangalore migrates with social capital: education, digital literacy, professional networks. He can organize formal associations, maintain websites, coordinate events at convention centers. The powerloom worker in Surat migrates with labor: his hands, his endurance, his caste network’s guarantee. He organizes around survival — a mess that feeds him, a dormitory that houses him, a festival that reminds him he is Odia. The bandwidth for political organization, legal advocacy, or institutional community-building is consumed by the twelve-hour shift.

Political Disenfranchisement

Here is the structural irony that defines the political existence of Surat’s Odia population: they are the single largest block in the city’s migrant workforce, and they have zero political voice.

Under Indian election law, a voter can only cast a ballot in the constituency where they are registered. Odia workers in Surat are registered to vote in Ganjam. To vote, they would have to travel 1,600 kilometers back to their home villages, lose days of work (for which there is no paid leave), and spend money they cannot afford — all for the privilege of participating in the democracy whose economy their labor sustains. The alternative — registering to vote in Surat — would require proof of local residence, documentation most workers do not have, and the willingness to sever the last formal tie to their home constituency.

The result is that they are politically invisible in both places. In Gujarat, political parties have “found nothing appealing to these power loom workers,” as The Wire documented. No candidate campaigns for the Odia vote in Surat. No party manifesto mentions powerloom worker welfare. The workers lack health insurance, provident fund, gratuity, or any occupational hazard relief — and no Gujarati politician has an incentive to address this, because these workers cannot vote in Gujarat.

In Ganjam, their absence distorts elections. Berhampur (the major city of Ganjam district) recorded only 63 percent turnout of 1.6 million voters in the 2024 elections — the lowest in the state, attributed to the mass absence of migrant workers. About 200,000 voters in western Odisha’s migration-prone districts are effectively disenfranchised in the same way. The irony deepened in 2024 when Naveen Patnaik, the twenty-four-year chief minister, chose Kantabanji — the hub of western Odisha’s dadan labor migration — as his second constituency. He lost by 16,344 votes to BJP’s Laxman Bagh. The BJP had used “Dadan Khati” — the mass departure of laborers under BJD’s watch — as a central campaign weapon.

Surveys suggest that about sixty percent of migrant workers nationally have missed voting at least once because they were away from home. For Surat’s Odia workers, the proportion is likely higher. A community of seven to eight lakh people — enough to fill multiple parliamentary constituencies — exercises approximately zero democratic power. They contribute to Surat’s GDP but not to Surat’s politics. They constitute Ganjam’s electorate but are absent from Ganjam’s elections. They are, in The India Forum’s phrase, “guest workers who became ghost workers.”

This is not a policy gap that can be fixed by extending postal ballots or setting up migrant voting booths, though both would help. The disenfranchisement is structural: a population that lives in one state and is registered in another, that has no formal employer to vouch for its existence, that works twelve hours a day without leave — this population is designed out of democracy by the intersection of election law, labor informality, and the sheer exhaustion of being a powerloom worker. The system does not need to actively suppress their vote. It only needs to make voting logistically impossible, which it does without any deliberate effort.

The 2006 Surat Floods

On August 7, 2006, the Ukai Dam on the Tapti River released water without adequate warning. The flood that followed submerged eighty to ninety-five percent of Surat. It was catastrophic for the entire city, but the geography of damage mapped with brutal precision onto the geography of Odia settlement.

The worst-hit areas were the low-lying industrial zones: Udhna, Limbayat, Ved Road — exactly the neighborhoods where Odia workers live and work. These workers occupied ground-floor and basement accommodations adjacent to powerloom units. They had no vehicles to flee. They had limited access to information about the dam release. They were, by the physical facts of where they lived and worked, maximally exposed.

Specific casualty and displacement data for the Odia community was not systematically collected. This is itself the story — a population large enough to constitute a small city, devastated by a flood that destroyed its living and working quarters, and no one counted the damage to them specifically. They were included in the aggregate Surat flood statistics the way they are included in everything: anonymously, as undifferentiated labor.

The flood exposed a structural vulnerability that COVID-19 would later expose at a different scale: migrant workers who have no formal existence in a city are also invisible in its disaster response. No registration means no targeted relief. No voter ID means no political constituency demanding recovery resources. No employer records means no liability. The flood came, destroyed, receded. The workers returned to rebuilt looms and the same rooms. The event left no policy trace specific to the migrant community.

COVID-19: The Mass Revelation

If the 2006 flood was a local disaster that the Odia community absorbed in silence, COVID-19 was a national crisis that made the community impossible to ignore.

The nationwide lockdown announced on March 24, 2020, shut down Surat’s powerloom industry overnight. Workers lost income instantly. They had no savings. They had no formal employment that might trigger any government relief mechanism. They could not buy food, could not pay rent, could not work. They were trapped — in a city that was not theirs, in rooms designed for sleeping between shifts that had suddenly become prisons.

What followed was not quiet suffering. On March 30, more than ninety Odia workers were arrested for defying the lockdown. In April, eighty-one Odia migrant workers were arrested for setting over ten vehicles ablaze in protest, telling police they were not getting food or basic facilities. In May, over a thousand migrant workers — many Odia — clashed with police, demanding to be sent home. Satya Swain, from Kullada village near Bhanjanagar in Ganjam, who had worked in Surat for two years, was allegedly beaten to death by police enforcing the lockdown.

The images that emerged — thousands of workers on highways, carrying their possessions on their heads, crowding railway platforms — became some of the defining visuals of India’s pandemic. And for many Indians, they were the first indication that a population the size of a mid-sized city existed in Surat, entirely composed of people from one Odia district, living in conditions that the word “precarious” does not begin to describe.

The return numbers told the statistical version of the story. By mid-June, 5,43,905 Odia migrants had returned by various transport modes. The Shramik Special trains alone brought 3,58,401 back to Odisha, with Gujarat contributing 36 percent — the largest share. When the Odisha government opened quarantine centers at the Gram Panchayat level, preparing 2.27 lakh beds across 7,200 isolation facilities, it was tacitly acknowledging a population it had not formally counted.

Almost ninety percent of Odisha’s 1,189 COVID cases as of late May 2020 were attributed to returning migrants. The case count rose from 162 on May 3 to 611 in just eleven days after migrant returns began. The migrants had not brought COVID to Odisha. They had brought evidence of their own existence.

“Almost everyone who had returned said they didn’t want to go back to the cities,” Mongabay reported. The desire was genuine. The economics were not. MGNREGA pays Rs 237 per day; even with the supplementary scheme for migration-prone blocks (Rs 352 per day), a worker would earn Rs 8,800 per month working twenty-five days — versus Rs 20,000 to Rs 25,000 in Surat. The arithmetic answered itself. “After almost five months of unemployment, workers have now started to travel back to Surat,” The Wire reported by late 2020. The pandemic had revealed the scale, exposed the conditions, generated national sympathy, and changed nothing. The wage differential remained. The looms needed operators. The fields in Ganjam still did not pay enough. The corridor reopened.

What COVID revealed, for those who looked carefully, was not just the size of the migration but its structural character. Four things became clear.

First, the true scale was far larger than any official estimate. Census 2011’s figure of 12.71 lakh interstate out-migrants from Odisha was massively outdated. When a million people registered on a single portal, the actual number of Odias living outside the state was clearly several multiples of the census figure.

Second, the informality was total. Workers had no ID cards, no registration, no employer records. The government struggled to locate and count them. A population that had been running India’s polyester industry for decades had left no formal trace of its existence.

Third, the dependency was mutual. When Odia workers left, Surat’s textile industry ground to a halt. This was not a one-way extraction where the workers needed Surat more than Surat needed them. The relationship was symbiotic — but the symbiosis operated on radically unequal terms.

Fourth, the vulnerability was not an aberration. Workers lost jobs instantly, had no savings, could not access services, had no social security — not because of the pandemic specifically, but because these protections had never existed. COVID did not create the precarity. It illuminated it.

The Second Generation

The corridor’s eighty-year history means that the second generation is not a future phenomenon. It is here.

Children born in Surat to Odia parents face the classic diaspora identity challenge, but in a peculiar form. They are not immigrants in a foreign country; they are internal migrants within their own nation, growing up in a city that does not count them as its own. Some attend Odia-medium schools — the Beena Odia Primary School and Divyajyoti Vidhyalaya — but most attend Gujarati or Hindi-medium schools. They speak Hindi and Gujarati at school and with friends. At home, they may speak Odia — if their parents maintain the language, if there is time between shifts, if the cultural weight of a language can survive when the entire external environment speaks something else.

No academic study specifically examines identity formation among second-generation Odia children in Surat. This absence is itself a data point. A community large enough to have its own schools is invisible enough to attract no scholarly attention. The broader literature on second-generation migrants in India suggests what we would expect: confusion at the crossroads of two cultures, gradual language erosion, a weakening connection to the place of origin that the first generation still calls home.

The marriage patterns tell a pragmatic story. Workers in Surat typically marry partners from Ganjam — the caste networks that structure employment also structure matrimony. But as the second generation grows, the question of whether marriages will continue to be arranged with partners from home, or whether the Surat-born generation will marry within Surat’s Odia community, will test the corridor’s social architecture.

The children are the corridor’s most revealing product. If they grow up speaking Odia, celebrating Nuakhai, marrying partners from Ganjam, and working the looms their parents worked — the corridor has successfully reproduced not just labor but community, for another generation. If they assimilate into Gujarati or Hindi-medium modernity, attend college, enter the formal economy, move beyond the powerloom — the corridor has served, despite everything, as an escalator. The data to know which outcome is more common does not yet exist. What we know is that the schools exist, the festivals continue, and the looms still need operators.

Why Surat and Not Bhubaneswar?

This is the question that every section of this chapter has been circling, and it has a simple answer and a complicated one.

The simple answer is wages. A powerloom worker in Surat earns Rs 20,000 to Rs 25,000 per month. There is no comparable labor-intensive industry in Odisha that pays anywhere near this. MGNREGA provides Rs 237 per day. Even the enhanced supplementary scheme in migration-prone blocks (Rs 352 per day) yields Rs 8,800 per month. The minimum wage for unskilled labor in Odisha is Rs 452 per day — roughly Rs 11,752 per month, still half of what Surat pays. And minimum wage enforcement in Odisha’s informal sector is, generously, aspirational.

The worker makes a rational calculation. Three times the income in Surat, even after subtracting Rs 2,500 per month for mess fees, means remittances of Rs 15,000 or more sent back to Ganjam. The Gram Vikas survey of Jagannathprasad block estimates annual remittances of Rs 64 crore from that single block. For the district as a whole, remittance estimates run to Rs 120-124 crore per month — roughly Rs 1,440 to Rs 1,488 crore per year, or approximately $170-175 million. From one district, to one city.

The complicated answer is the one that matters for policy. Why doesn’t Odisha have industries that employ its own people at comparable wages?

The state has tried. Odisha TEX 2025, the textile investment summit, generated thirty-three MoUs worth Rs 7,808 crore, projecting 53,300 direct jobs with worker subsidies of Rs 6,000 per month (male) and Rs 7,000 per month (female). The target is one lakh textile jobs by 2030. Sisir Gouda, a Ganjam man who worked in Surat and Mumbai textile mills for thirty-two years, returned in 2020 to set up Matexmate Textile Private Limited in his native village, investing Rs 2 crore and installing nine weaving machines with capacity for fifty.

But here is the sobering math. Even if Odisha TEX’s full ambition materializes — one lakh jobs — it would absorb roughly one-seventh of the Odia workforce currently in Surat. And the track record of investment summits materializing in Odisha is, to put it carefully, mixed. The gap between announcement and functioning factory is measured not in months but in years, sometimes decades. Quora discussions among educated Odias cite a figure that, even if imprecise, captures a pattern: Rs 67,000 crore in announced FDI “commitments” versus Rs 39 crore in actual FDI in a recent year.

The reason is not that Odisha lacks comparative advantage in textiles. It has the labor force — they are already trained, in Surat. It has the coast — for raw material import and finished goods export. It has lower land costs than Gujarat. What it lacks is the ecosystem: the cluster of suppliers, the logistics infrastructure, the electricity reliability, the regulatory environment, the accumulated knowledge of running a textile production center that Surat has built over fifty years.

In software engineering terms, Surat has network effects. The powerloom industry is not a collection of independent factories; it is a cluster — with shared supply chains, specialized repair services, a fabric market that is one of Asia’s largest, dyeing and processing units, a transportation network, a labor recruitment system. Replicating the output of a single factory is straightforward. Replicating the ecosystem is a different problem entirely, requiring coordinated investment across dozens of interdependent capabilities over a sustained period. No Indian state has done this quickly. The ones that have done it at all — Tamil Nadu for automobiles, Gujarat for chemicals, Karnataka for IT — took decades and benefited from specific historical contingencies that cannot be manufactured on demand.

This is not an argument for resignation. It is an argument for honesty about what the problem actually is. The Surat corridor will not be reversed by a textile investment summit or a worker subsidy scheme, though both are useful. It will be reversed — if it is ever reversed — by the slow, unglamorous work of building an industrial ecosystem in Odisha that creates enough demand for labor, at sufficient wages, that the rational calculation flips. That is a project measured in decades, not electoral cycles. And it requires something that neither the BJD’s twenty-four-year government nor the BJP’s new administration has yet demonstrated: the capacity to turn announced investment into functioning, employing, wage-paying industry at scale.

A Permanent Transplantation

The Surat corridor is not a temporary phenomenon. This is worth stating plainly, because much of the policy discourse around migration treats it as a problem to be solved — as though, with the right scheme, the right investment, the right amount of MGNREGA days, the workers will come home and the corridor will close.

It will not. The corridor is eighty years old. It spans four generations. It has its own schools, its own mess economy, its own caste structure, its own festival calendar. It survived a catastrophic flood. It survived a pandemic. When COVID emptied Surat of its Odia workers, the workers returned within months, because the fundamental equation had not changed.

The corridor is, in fact, a form of state-building — though not the kind any government intended. Seven lakh Odias in Surat have created, without planning or policy, a functioning community with internal governance (the mess system), an economic base (the powerloom industry), social services (the community organizations and schools), cultural institutions (the festival celebrations), and a financial system (remittance flows that sustain an entire district’s economy). What they have not created, because the structure prevents it, is political representation.

The honest assessment is this: Odisha’s Surat corridor is a permanent transplantation of a community from one Indian state to another. It is not a migration crisis to be solved. It is a structural feature of the Indian economy, in which labor moves to where capital has clustered, and the social institutions that make that labor possible follow it. The corridor will continue as long as Surat’s textile industry needs cheap, disciplined labor and Ganjam cannot offer an alternative. The workers are not waiting for a government scheme to bring them home. They are building lives — constrained, difficult, often exploitative lives — in the city that pays them.

What the corridor asks, to anyone willing to hear the question, is not “how do we bring them back?” but something harder: why, after eighty years, does India’s most mineral-rich state still export its people as its most reliable product? The looms run. The fabric rolls off. The money flows back to Ganjam. The villages build pucca houses with remittance money and empty rooms with absent men. The trains from Udhna Junction to Berhampur run full in both directions. And the corridor, older than independent India’s industrial policy, continues — indifferent to elections, immune to schemes, sustained by the most durable infrastructure Indian society has ever built: the networks of caste and kinship that connect a southern Odisha village to a Gujarati industrial city, as reliably as any road the government has failed to complete.

Sources

This chapter draws on research compiled in three documents:

reference/the-leaving/odisha-migration-statistics-research.md — Census 2011 data, OMS 2023, COVID-19 migration data, push factors, government response (75 sources)
reference/the-leaving/odisha-diaspora-social-cultural-research.md — Surat corridor history, powerloom structure, caste dynamics, living conditions, cultural life, political exclusion (100+ sources)
reference/the-leaving/odia-diaspora-online-discourse-research.md — Diaspora voices across digital platforms (50+ sources)

Key investigative sources:

Source Research

The raw research that informs this series.