Chapter 2: The Dadan Road

---

In May 2020, as India’s COVID-19 lockdown entered its third month, a nineteen-year-old woman named Manasi Bariha made a phone call that would trigger one of the largest bonded labor rescues in modern Indian history. Bariha was from Balangir district in western Odisha, one of hundreds of thousands of Odias who travel each year to work at brick kilns across southern India. She had been trapped with 360 other workers at a kiln in Pudhukuppam, Tamil Nadu. The lockdown had shut down the kiln, but the owner would not release the workers. No wages. No food supply. No permission to leave. Bariha managed to contact journalists and the International Justice Mission, an anti-trafficking organization. Within forty-eight hours of the raid that freed her group, Indian law enforcement swept through thirty-one kilns in the surrounding area. They freed 6,750 people.

Six thousand seven hundred and fifty. Not from a war zone. Not from a collapsed state. From brick kilns within a few hours’ drive of Chennai, one of India’s most economically advanced metropolitan areas, in the year 2020. The workers had been making bricks — the literal building blocks of India’s construction boom — under conditions that met every legal and scholarly definition of modern slavery: debt bondage, restriction of movement, confiscated identity documents, wages set below subsistence, violence for non-compliance.

Manasi Bariha’s phone call revealed what the system normally keeps invisible. The dadan road — the annual migration of western Odisha’s poorest people into bondage at brick kilns, construction sites, and factories across India — operates in a zone of engineered obscurity. The kilns are in remote locations. The contracts are verbal. The workers are illiterate and without documents. The journeys happen in October and November, when nobody is watching, and the returns happen in May and June, if they happen at all. The system has been documented since at least the 1970s. It has not been solved. It has, by most measures, grown.

This chapter follows the dadan road from beginning to end — the recruitment, the advance, the journey, the work, the trap, and the economics that keep the whole apparatus turning. The question is not whether it is unjust. That much is obvious. The question is why it persists, and what that persistence reveals about the structure of Odisha’s economy and India’s labor markets.

---

What Dadan Means

The word itself is the first thing to understand. “Dadan” traces to the Persian “dadni,” meaning an advance payment. In Odia, it has acquired a specific and darker meaning: migration through debt. The word does not merely describe a financial transaction. It describes a system — one in which the advance payment is not a benefit extended to the worker but a trap constructed around them.

In software engineering, there is a concept called “technical debt” — shortcuts taken during development that make immediate progress possible but accumulate costs that must eventually be paid, often at compound interest. The dadan advance operates on the same principle, except the interest compounds against human bodies. The worker takes the advance because they need cash immediately — to repay a moneylender, to buy seed, to survive the monsoon, to celebrate Nuakhai. The advance solves the immediate crisis. But it creates an obligation that the worker cannot discharge through labor at the rates they will be paid, in the time they have been allotted. The debt carries forward. The cycle repeats. Technical debt, if left unaddressed long enough, makes a software system unmaintainable. Dadan debt, left unaddressed across generations, makes a family’s economic independence unachievable.

The system has a formal name in academic literature: “advance-linked seasonal labor migration.” But the people who live inside it call it dadan, and in Balangir they call the acceptance of the advance “bahu bandha” — the tying of arms. The metaphor is precise. Once your arms are tied, you work where you are told to work.

---

How the System Works

The dadan cycle follows a calendar as regular as agriculture, because it is agriculture’s mirror image — it fills the economic vacuum that failed agriculture leaves behind.

Step one: The approach. Between August and October, during and after the Nuakhai harvest festival, labor contractors called sardars visit villages in western Odisha’s migration-prone districts — Balangir, Nuapada, Kalahandi, Bargarh, Sonepur, Boudh. The timing is deliberate. Nuakhai is the most important festival in western Odisha. Families need cash for new clothes, for feasting, for the social obligations that hold communities together. The monsoon crop is in, but yields on unirrigated land in Balangir — where only three percent of agricultural land has irrigation — may have been poor. Debts from last year’s lean season are due. The sardar arrives when the need is sharpest.

He does not advertise. He does not need to. He is often from the same village or the same caste. He may be a former migrant himself who graduated from laborer to recruiter — a promotion that pays better and requires no bricks. He knows which families are in debt, which households had crop failures, which daughters have weddings approaching. He targets the poorest: landless Dalits, Adivasis, marginal farmers whose kharif harvest will not feed a family through the winter.

Step two: The advance. The sardar offers cash. The amounts have risen over time — Rs 10,000 to Rs 15,000 per individual worker for basic brick kiln labor, Rs 40,000 to Rs 60,000 per family unit (typically husband and wife, sometimes children), and up to Rs 1 lakh in some recent cases. The advance is higher for stronger workers, for families with more working members, for people who have been reliable in previous seasons. This is not generosity. It is pricing. The advance is calibrated to be large enough that the worker cannot refuse it and cannot repay it from any local source.

There is no written contract. The agreement is verbal. The sardar promises wages, accommodation, and food at the destination. The worker pledges six to eight months of labor, from November through May or June. The promise is specific enough to recruit and vague enough to be unenforceable. Nothing is signed. No copies are kept. No government office is informed.

Step three: The network above. The sardar is the visible face of a layered system. Above him is the thikadar — a higher-level contractor who manages the logistics of transport, negotiates with brick kiln owners, and oversees multiple sardars. Above the thikadar is the kiln owner or factory operator. Each layer takes a cut. The sardar earns a commission per worker recruited plus a percentage of production. The thikadar takes a margin between what the employer pays and what he passes down. The employer gets a workforce that is captive, cheap, and has no bargaining power. The spread between what the employer pays per worker and what the worker actually receives can be enormous, but because the accounting is opaque and the workers are illiterate, nobody at the bottom of the chain knows the actual numbers.

Think of this as a supply chain, because that is what it is. The product being moved is human labor. The sardars are distributors. The thikadars are logistics providers. The kiln owners are end customers. The workers are the raw material. Every intermediary in the chain extracts value, and the person who creates the value — the worker who actually makes the bricks — receives the smallest share. This is not unique to the dadan system. It is how many supply chains work. What makes dadan different is that the raw material cannot walk away, because it owes money.

Step four: The journey. In October and November, the workers leave. They travel by train from stations like Kantabanji and Khariar Road in Balangir district — stations that become, during these months, some of the busiest departure points for labor migration in India. Some travel by truck, packed into cargo beds. Entire families go: husband, wife, children. The children are not being taken on a trip. They will work.

The destinations are distant by design. Brick kilns in Andhra Pradesh, Telangana, Tamil Nadu, Karnataka. Construction sites in Hyderabad, Chennai, Bangalore. The distance serves a function: it makes escape expensive and disorienting. A family from a tribal village in Nuapada, transported to a kiln in a remote part of Tamil Nadu, does not know the local language, does not know the geography, has no local contacts, and has no money for a return ticket. The distance is not incidental. It is infrastructure.

Step five: The documents. At some point during the recruitment or the journey, the sardar collects the workers’ Aadhaar cards, ration cards, and any other identity documents. The stated reason is “safekeeping.” The actual function is control. A worker without identity documents in a state where they do not speak the language and have no legal address is, for practical purposes, a person who does not exist. They cannot approach police. They cannot access government services. They cannot prove who they are. Document confiscation is one of the internationally recognized indicators of human trafficking, and it is routine in the dadan system.

Step six: The work. And then the bricks begin.

---

The Brick Kilns

Brick kilns are the largest single destination for dadan labor from Odisha. They are also the most documented, because the conditions in them are extreme enough to attract the attention of journalists, NGOs, and occasionally law enforcement.

The geography is specific. The kilns are in the hinterlands of Andhra Pradesh, Telangana, Tamil Nadu, and Karnataka — states that need bricks for their construction booms but do not have enough local labor willing to work at kiln rates. The kilns are located in “interior and deserted locations,” as the International Justice Mission describes them — far from towns, far from roads, far from anyone who might ask questions. The remoteness is, again, functional. A kiln in a busy town would attract labor inspectors. A kiln fifteen kilometers down an unpaved road from the nearest village does not.

The work is piece-rate. A family unit is assigned a production target — a common one is six lakh bricks (600,000) over the six-month season. The family molds, stacks, dries, and loads clay bricks, fifteen or more hours per day, seven days per week. There is no concept of a weekly rest day. There is no concept of sick leave. “We could hardly sleep for three to four hours a day,” one brick kiln worker named Kumudini told The Federal. “The owner and his men would abuse us with filthy language.”

The physical conditions are what you would expect in an operation optimized for output and indifferent to the input material. Shelter is makeshift — tents or lean-tos near the kiln. Water supply is inadequate. Only one in five workers has access to a washroom. There are no medical facilities. The kiln itself radiates heat that can make the surrounding area unbearable, and the workers are in it from before dawn until after dark.

The workforce is young. Seventy-two percent of dadan brick kiln workers are between sixteen and twenty-nine years old. This is not because brick-making is a young person’s career choice. It is because the work is so physically destructive that older bodies cannot sustain it. By the time a worker reaches their thirties, they are often too damaged to continue. This is labor that consumes people at a rate faster than it pays them.

---

The Wage Trap

Here is where the system’s architecture becomes visible.

During the season, the kiln owner does not pay full wages. Workers receive a small food allowance — enough to survive, not enough to save. The actual wage settlement happens at the end of the season, in May or June, when the monsoon rains make brick-making impossible. At settlement time, the worker discovers what they have “earned.”

The accounting is opaque by design. The advance given by the sardar is deducted. Food allowances are deducted. Any “expenses” the kiln owner claims to have incurred — transport, materials, equipment — are deducted. Penalties for days missed due to illness or injury are deducted. In the employer’s books, money siphoned by intermediaries is often listed as the worker’s debt. The worker, who cannot read the ledger and has no independent record of their labor, must accept whatever number the owner states.

The result is predictable. Many workers discover at the end of the season that they have not cleared their advance. They still owe money. The debt carries forward into the next year, and the cycle begins again — except now the worker returns to the village with nothing to show for six months of labor except a body more worn than when they left.

For those who do clear their advance and return with some cash, the economics of the off-season conspire to restart the cycle. The monsoon months — June through September — are the period when brick kilns do not operate, when agricultural work in western Odisha is limited (remember: three percent irrigation in Balangir), and when families must survive on whatever they have saved. The savings, if any, run out. The moneylenders charge high interest. The next Nuakhai approaches, and with it the social obligation to celebrate, to feast, to buy new clothes. The sardar reappears. The advance is offered again. The arms are tied again.

This is not a failure of individual decision-making. It is a system designed so that the rational choice at every decision point — accept the advance or starve, stay at the kiln or lose the advance, return next year or face debts with no income — leads to the same outcome: another season of bondage. Game theorists would recognize this as a repeated game in which one player sets the rules, controls the information, and can punish defection, while the other player’s only alternative to playing is destitution. The equilibrium is bondage. It is stable. It is, from the employer’s perspective, efficient.

---

Who the Sardars Are

The sardar is the most psychologically complex figure in the dadan system, and understanding him — they are almost always men — is essential to understanding why the system resists reform.

The sardar is typically from the same community as the workers he recruits. He may be from the same village. He speaks the same language, belongs to the same caste or a nearby one, worships at the same temple. He is not an outsider imposing exploitation from without. He is a neighbor facilitating it from within. This is the exploiter-as-neighbor paradox, and it is one of the dadan system’s most effective structural features.

Many sardars are former migrants. They worked the kilns themselves, learned the business, built relationships with thikadars and kiln owners, and eventually shifted from selling their own labor to selling other people’s. The transition from worker to recruiter is a kind of promotion — the only upward mobility the system offers. It pays better, it is less physically destructive, and it confers a form of local power. The sardar who can advance cash to a desperate family commands a social authority that the village schoolteacher or the MGNREGA supervisor does not.

This creates a structural problem for any reform effort. The sardar is simultaneously the person who exploits and the person who provides. When no bank will lend to a landless tribal family in Nuapada, the sardar will. When no government program delivers employment during the lean season, the sardar offers work. When a family faces a medical emergency and needs Rs 50,000 immediately, the sardar can produce cash within days. He fills a vacuum that formal institutions have left empty. Removing the sardar without filling the vacuum does not liberate the worker. It removes their only source of credit and employment, however exploitative.

In some cases, the person who should be protecting workers from the dadan system is the person running it. Local sarpanches — elected village heads — sometimes double as labor contractors. This creates a direct conflict of interest. A sarpanch who profits from labor recruitment has no incentive to ensure that MGNREGA works are available during the lean season, because MGNREGA employment competes with his recruitment pipeline. The institution designed to serve the poor is, in these cases, captured by the system that exploits them. This is not an anomaly. It is a predictable outcome when formal institutions are weak and informal power networks are strong.

---

Beyond the Kilns: Construction and Other Destinations

Brick kilns receive the most attention because their conditions are the most visibly extreme, but they are not the only destination on the dadan road. Construction sites across India’s expanding cities — Hyderabad, Chennai, Bangalore, Mumbai — absorb large numbers of western Odisha migrants through similar contractor-mediated systems.

Construction migration is less documented than kiln migration, for a reason that reveals something about how we pay attention to suffering. Brick kilns concentrate workers in a single remote location, which makes them findable when raids happen. Construction sites disperse workers across a metropolitan area, each site employing a handful of migrants among larger workforces. The exploitation is the same — no formal contracts, piece-rate wages, no safety equipment, exposure to dust and chemical hazards, no medical care — but it is distributed thinly enough across enough sites that no single instance generates the dramatic rescue numbers that make headlines.

Workers from Balangir and Kalahandi have been documented on construction sites in Hyderabad. Migrants from Gajapati and Koraput go to construction in Kerala and Tamil Nadu. Domestic work absorbs women, particularly from KBK districts, into Delhi and Kolkata households where they are essentially invisible. The unorganized sector — which is to say, the sector where labor laws do not effectively operate — absorbs Odia migrants like water absorbs salt: completely, and with no visible trace.

The total scale across all destination industries is difficult to pin down, for reasons that are themselves part of the story. These workers are not registered. They do not appear in destination-state labor records. They are classified, if they are classified at all, as casual workers even after ten or fifteen years in the same industry. The formal economy does not see them. The informal economy could not function without them.

---

What Happens When Things Go Wrong

The documented atrocities in the dadan system are not outliers. They are the logical consequences of a labor arrangement in which workers have no legal standing, no physical freedom, no bargaining power, and no one watching.

In one widely reported case, a middleman who recruited workers for brick kilns in Andhra Pradesh chopped off the hands of workers who refused to work. ActionAid, the international advocacy organization, called it a “savage act” and demanded prosecution. The description is accurate. But the act was not random savagery. It was enforcement — the same logic that leads any system built on captive labor to use violence to maintain compliance. When a worker’s only value is their labor, and their labor is secured by coercion rather than contract, the coercion must escalate when the labor is refused. The hand-chopping is the system working as designed.

A couple from Bolangir employed at a brick kiln in Kurnool, Andhra Pradesh, was found dead in a hospital. Injury marks covered their bodies. The palms of both hands had been chopped. Whether this was punishment, murder to prevent testimony, or something else was not established with certainty. Certainty requires investigation, and investigation requires someone with authority to care.

Seventy-two people, including seven girls, were rescued from a brick kiln in Sangareddy, Telangana. The girls were admitted to a Sakhi centre, and the owners were booked under the POCSO Act (Protection of Children from Sexual Offences) and the Child Labour Act. Sexual abuse of women and children at brick kilns is documented often enough that its absence from most reporting likely reflects under-reporting rather than under-occurrence.

On February 6, 2024, news broke that a migrant worker’s family from Bolangir had their younger daughter kept “under mortgage” at a brick kiln in Andhra Pradesh — held as collateral for their parents’ debt. The child was not working off a debt. She was the debt’s guarantee: her body held hostage to ensure her parents’ continued labor.

Between 2010 and 2015, 1,208 laborers — including 500 women and 100 children — were legally freed as bonded laborers from brick kilns in Tamil Nadu, Andhra Pradesh, and Karnataka. NGOs like Samata and the KBK Resource Centre rescued more than 10,000 bonded laborers, including child laborers, from brick kilns in Andhra Pradesh and Telangana over a four-year period. In 2024, 626 workers from Balangir and 153 from Nuapada were rescued. In 2025, 572 from Balangir and 134 from Nuapada. These are the ones who were found. The ones who were not found continued making bricks.

---

Children on the Dadan Road

More than twenty percent of workers rescued from brick kilns between 2011 and 2013 were children. This number requires a moment of attention, because it is easy to read past it.

One in five. Children between six and fourteen, working twelve to fifteen hours daily, molding and carrying bricks, in conditions that damage adult bodies within a few seasons.

They are there because their parents face an impossible choice. Leave the children in the village with elderly grandparents — in a place with poor schools, no supervision, no working-age adults — or take them to the kiln where they will work instead of study. Both options damage the next generation. The damage just takes different forms. Left behind, the children face neglect and early engagement in economic labor. Taken along, they face malnutrition, stunted growth, interrupted education, and workdays that no child should endure.

“If we do not take our children along, how can we repay?” — a bonded laborer explaining the logic. The logic is sound. A family assigned a production target of 600,000 bricks in six months needs every pair of hands. The children are not excess labor. They are necessary labor. The family cannot meet the target without them, and failing to meet the target means the advance is not repaid, which means the debt carries forward, which means another season, which means the children come again next year too.

Odisha’s school dropout rate rose from twelve percent to fifteen percent in 2024-25. At the secondary level, the state’s dropout rate is 27.3 percent — more than double the national average of 12.6 percent. Migration is explicitly cited as a major driver. Children who drop out because their families migrate have lower skills, which limits them to the same informal labor their parents do. The intergenerational trap closes. The dadan system does not just exploit the current generation. It recruits the next one.

---

The Law on Paper

India does not lack laws against bonded labor and the exploitation of migrant workers. It lacks enforcement.

The Bonded Labour System (Abolition) Act was passed in 1976 — the year after Odisha passed its own Dadan Labour (Control and Regulation) Act of 1975, making Odisha the first state in India to formulate legislation specifically protecting migrant workers. Four years later, the national government passed the Interstate Migrant Workmen (Regulation of Employment and Conditions of Service) Act of 1979. The 1979 Act requires any establishment employing five or more interstate migrant workers to register with the government. It mandates written contracts, prescribes minimum wages, and requires employers to provide housing, medical facilities, and displacement allowances. In 2020, the Occupational Safety, Health and Working Conditions Code subsumed the 1979 Act with similar provisions.

On paper, a brick kiln in Tamil Nadu employing two hundred workers from Odisha must register them, provide written contracts, pay minimum wages, furnish housing and medical care, and comply with regular inspections. On the ground, none of this happens.

The Observer Research Foundation described the Interstate Migrant Workmen Act as “a dead letter” in 2020, and the description is not rhetorical. A dead letter is a law that exists in legislation but not in practice — a legal fiction that provides the appearance of protection without its substance. The 1979 Act has been a dead letter for its entire existence. No state systematically enforces registration of contractors or establishments employing migrant workers. District Labour Officers in Odisha are designated as Registering Officers under the Act, but their enforcement capacity is negligible. Brick kiln owners operate in remote locations precisely to evade detection. Workers lack awareness of their legal rights, and sardars deliberately keep them uninformed.

The structural problem is fundamental: the law assumes a formal employer-employee relationship. The dadan system is informal by design. Every feature of the system — verbal contracts, cash advances, no documentation, remote worksites, illiterate workers, confiscated identity papers — is optimized to make the law inapplicable. The system did not evolve to evade the law accidentally. It evolved in the law’s absence and is sustained by its irrelevance.

There is a concept in cybersecurity called an “air gap” — a physical separation between a secure network and an unsecured one that makes electronic intrusion impossible. The dadan system maintains an air gap between the legal framework and the labor market. The laws exist in the world of formal employment, written records, and institutional enforcement. The dadan market operates in the world of verbal agreements, cash, and personal power. The two worlds do not touch. An inspector would need to find a kiln in a remote location, gain access despite the owner’s resistance, identify workers who are terrified of retaliation, communicate in a language the workers understand, and then pursue prosecution through a court system in a state where neither the workers nor their families live. The practical barriers are not accidental. They are the system’s immune response.

Why are prosecutions rare? Because prosecution requires a victim who will testify. A worker who has been freed from a kiln and returned to their village in Odisha has little incentive and no practical means to travel back to Tamil Nadu or Andhra Pradesh to appear in court. The worker needs to eat, and the next lean season is approaching. The rescued workers from Manasi Bariha’s kiln were freed in May 2020. How many of them were in a courtroom in Tamil Nadu by 2021? How many were, instead, back at another kiln, because the alternative was worse?

---

The Economics of Persistence

The dadan system persists because it solves a problem for every participant — except the worker, for whom it creates a different problem.

For the worker: Even exploitative wages at a brick kiln exceed what is available at home. MGNREGA, the national rural employment guarantee, pays Rs 237 per day in Odisha. Even with the state’s supplementary scheme that adds Rs 115 per day, the total is Rs 352 — and that assumes MGNREGA work is actually available, which in many migration-prone blocks it is not. Average employment generated per household under MGNREGA in 2023-24 was 53.23 days. Fifty-three days of work at Rs 352 per day is Rs 18,752 for the year.

A dadan advance alone can be Rs 40,000 to Rs 60,000 — more than what MGNREGA delivers in a year, provided in a single payment, in cash, when it is needed most. The fact that the advance is a trap rather than a gift does not change the immediate calculus. When your daughter’s wedding costs more than your annual income, and the sardar is offering two years’ worth of MGNREGA wages in his hand, the rational choice is to take the money. Rationality under constraint looks different from rationality under abundance.

At the kiln, even the exploitative wage — Rs 8,000 to Rs 12,000 per month, often paid only as a food allowance during the season — compares favorably to what is available in Balangir during the lean months, which is close to nothing. A Surat powerloom worker earns Rs 20,000 to Rs 25,000 per month — roughly three to four times the MGNREGA rate. The wage differential is the fundamental engine of migration, and it exists because Odisha’s rural economy does not generate enough productive employment at competitive wages.

For the sardar: The commission system makes labor contracting profitable. Each worker recruited generates income, and the sardar who manages a hundred workers earns a reasonable living without touching clay. The sardar’s economics are straightforward: he is a middleman in a market with excess supply (desperate workers) and steady demand (kiln owners needing labor), and he captures the margin between them.

For the employer: Dadan labor is the cheapest, most controllable workforce available. Kiln owners could, in theory, hire local workers in Andhra Pradesh or Tamil Nadu, but local workers have options — they can leave for better jobs, they know their rights, they speak the local language, they have family nearby. A dadan worker from Odisha has none of these advantages. They are isolated, indebted, document-less, and linguistically stranded. They will accept conditions that no local worker would tolerate. This is not a labor market. It is a market in captive labor, and captive labor is cheaper than free labor because it cannot bargain.

For the state government: Out of sight, out of mind. Workers who migrate to other states are no longer visible in Odisha’s economic statistics. They are not counted as unemployed (they are working, technically). Their suffering happens in a different state’s jurisdiction. And critically, they do not vote in their destination state — but they also often do not vote at home, because they are absent during elections. An estimated 200,000 voters in western Odisha are effectively disenfranchised by migration. Berhampur in Ganjam recorded only sixty-three percent turnout among 1.6 million voters in the 2024 elections — the lowest, attributed to migrant absence. The people most harmed by government failure are precisely the people who cannot hold the government accountable, because they are in another state making bricks.

This is a stable equilibrium. Every participant’s incentives are aligned to perpetuate the system. The worker needs cash. The sardar needs commissions. The employer needs cheap labor. The government needs the problem to be invisible. The only force that disrupts the equilibrium is external intervention — raids, rescues, investigative journalism, activist litigation. And external interventions, by definition, are sporadic. The system is continuous.

---

The Scale of What We Are Talking About

The numbers are imprecise, because precision would require the kind of data infrastructure that does not exist for populations this invisible. But the available estimates, assembled from government tracking, NGO surveys, academic studies, and the involuntary audit that COVID-19 performed, give a sense of scale.

Over 60,000 families — approximately 200,000 individuals — from Balangir, Nuapada, Kalahandi, Boudh, Sonepur, and Bargarh migrate annually to brick kilns in other states. An additional 40,000 to 50,000 people move to brick kilns near Cuttack and Bhubaneswar within Odisha. Another roughly 100,000 migrate in unregistered, shadowy ways to construction sites and other informal-sector destinations that nobody counts.

Government tracking, which captures only a fraction of actual migration, showed 70,142 registered migrant workers in 2024 and 94,106 in 2025 — a thirty-four percent increase in a single year. Whether this reflects more migration or better tracking is unclear. Both are probably true. The trend line, in either case, goes up.

In Sargul village in Muribahal block of Balangir, ninety percent of the approximately 2,300 residents are in the dadan category. In Ichhapada panchayat, eighty percent. These are not statistics. They are descriptions of places where, between November and May, essentially no working-age adults remain. The schools have fewer children. The fields lie unworked. The homes are locked. Journalists who visit during the migration months consistently describe ghost-village conditions.

Over the past decade, 403 migrant workers from Odisha died in other states — a figure reported by the state’s Labour Minister in the Assembly. Ganjam reported the highest deaths at 59, followed by Kalahandi at 39 and Balangir at 35. The annual average is approximately sixty deaths per year. In just Surat’s registered powerloom units between 2012 and 2015, there were eighty-four fatal events killing 114 workers and seriously injuring 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).

These are the counted dead. The uncounted dead — workers who died at remote kilns, whose deaths were not reported, whose families learned of their loss through a phone call from a fellow worker rather than from any official channel — do not appear in any database. We do not know how many there are. We should be honest about that gap.

And then there are the 5,613 migrant workers rescued from distressed conditions at worksites over the past five years, as reported by the Odisha government. The 10,000-plus bonded laborers rescued by NGOs over a four-year period. These numbers represent successful interventions. They do not represent the universe of people who needed intervention and did not receive it. The rescued are a sample of the trapped. The sample size is large. What it implies about the population is larger.

---

Why Dadan Has Not Been Solved

In 2024, the BJP used “dadan khati” — the suffering of laborers forced to leave the state — as a political weapon against the BJD’s twenty-four-year rule. The charge was effective enough that it contributed to the BJD’s electoral defeat. Naveen Patnaik himself contested from Kantabanji, the western Odisha town that functions as one of the largest labor departure points in the country, as his second constituency. He lost there to the BJP’s Laxman Bagh by 16,344 votes.

The new BJP government, under Chief Minister Mohan Charan Majhi, constituted a twenty-member Distress Migration Task Force in October 2024, led by Deputy Chief Minister K.V. Singh Deo. The government extended MGNREGA to 300 days per year in thirty migration-prone blocks across nine districts. It launched awareness campaigns. It sent officials to destination states for rehabilitation planning. It signed thirty-three MoUs worth Rs 7,808 crore at Odisha TEX 2025, projecting 53,300 jobs in the textile sector.

These are real interventions. They are also insufficient, and the history of Odisha’s engagement with the dadan problem explains why.

The problem is structural, not programmatic. The dadan system exists because of a specific configuration of failures: poverty so deep that any cash advance is irresistible; agriculture so unproductive that the lean season offers no income; irrigation so absent that farming cannot be year-round; local industry so nonexistent that there are no jobs to compete with the sardar’s offer; governance so weak that laws on the books do not operate in the field; and political incentives so misaligned that the people who suffer most have the least electoral voice.

MGNREGA at Rs 352 per day — even for 300 days — is roughly Rs 1,05,600 per year. A family of four working at a brick kiln for six months, even at exploitative rates, earns more and receives a lump-sum advance that no government program provides. The wage differential has narrowed, but it has not closed. And MGNREGA does not solve the cash-flow problem: the advance is needed in September, not spread across twelve months. Until a worker can access Rs 40,000 to Rs 60,000 in cash during the lean season from a source that does not charge interest or demand bonded labor in return, the sardar’s offer remains the best available option.

Enforcement of labor laws would require cooperation between Odisha (the source state) and Andhra Pradesh, Telangana, Tamil Nadu, and Karnataka (the destination states). The source state has the incentive — its citizens are being exploited — but not the jurisdiction. The destination states have the jurisdiction but not the incentive — the exploited workers are building their construction sectors at below-market cost. Interstate enforcement coordination on labor matters is, in practice, nearly nonexistent. The Occupational Safety, Health and Working Conditions Code of 2020 was supposed to streamline interstate protections. It has not.

Addressing the root cause — making it economically viable for people to stay in western Odisha — requires irrigation, industrialization, financial inclusion, and institutional strengthening, none of which can be delivered by a task force or a single government scheme. Balangir’s three percent irrigation rate is not a statistic that changes in a five-year plan. It is the result of decades of underinvestment in water infrastructure. The absence of any significant manufacturing in the KBK districts is not a gap that a textile MoU fills — MoUs are announcements, not factories, and Odisha’s history is littered with announced investments (POSCO’s Rs 51,000-crore steel plant, ArcelorMittal’s project) that never materialized.

NewsClick described the Balangir task force as “a gold filling to hide decay” — cosmetic intervention masking deep structural failure. The description may be uncharitable. The interventions are real. But they are proportioned to the political cycle, not to the structural problem. A problem fifty years in the making — Odisha’s Dadan Labour Act was passed in 1975, half a century ago — will not be solved by a committee constituted after an election.

---

The System as Market

Here is the uncomfortable conclusion, and it is worth stating plainly: the dadan system is not an aberration in the market. It is a market response to a structural deficit.

Western Odisha has a surplus of labor and a deficit of employment. Southern India has a surplus of construction demand and a deficit of labor willing to work at the wages and conditions that maximize kiln owners’ profits. The dadan system connects supply to demand. It does so through exploitation, bondage, and coercion — but it does so because no other mechanism connects them at all. The formal labor market does not reach into Nuapada villages. The government employment guarantee does not fill the lean season. The banking system does not lend to landless Dalits.

The sardar fills every gap. He is the credit provider, the employment exchange, the travel agent, and the social contact in a distant state. He does all of this while extracting value from the people he serves, and the people he serves accept the extraction because the alternative is not a better sardar. The alternative is no sardar — which means no advance, no work, and no income during the eight months when agriculture provides nothing.

You do not solve the dadan system by banning the sardar. You solve it by making the sardar unnecessary. That means irrigation so farming generates year-round income. It means local industry so workers have options within their district. It means banking so families can access credit without pledging their labor. It means functional MGNREGA that pays competitive wages and actually delivers 300 days of work. It means schools that work, so the next generation has options this one did not.

The gap between what exists and what would need to exist is not a failure of will. It is a measure of how much has not been built. The dadan road will remain open as long as the road home leads nowhere.

---

What We Do Not Know

Honesty about the limits of available knowledge is a form of respect for the subject. There is much we do not know about the dadan system, and the gaps are not random — they correspond to the system’s designed invisibility.

We do not know the exact number of people in the dadan stream. Estimates range from 200,000 to 300,000 annually from western Odisha alone, but “annually” obscures the fact that many are semi-permanent migrants who return only briefly, if at all. The government tracking figure of 94,106 registered workers in 2025 is acknowledged by everyone, including the government, to be a fraction of the actual number.

We do not know how many workers die at kilns and construction sites without being counted. The official figure of 403 deaths over a decade is almost certainly an undercount, because it relies on deaths being reported through official channels — and the entire point of the dadan system is to operate outside official channels.

We do not know how many children are working at kilns right now. The twenty percent figure from 2011-2013 rescues is a decade old. The number could be higher or lower. Nobody is systematically counting.

We do not know what happens to workers after they are rescued. Village Square reported that rescued bonded laborers “languish” after being freed — the rescue removes them from the kiln but does not provide the economic alternative that would prevent them from returning to the next sardar who offers an advance. Rescue without rehabilitation is, in many cases, a pause in the cycle rather than an end to it.

We do not know whether the dadan system is growing or shrinking. The thirty-four percent increase in tracked migrants between 2024 and 2025 could reflect growth, or it could reflect better tracking, or both. The COVID-19 return data — in which over five lakh Odia migrants registered on a single portal — suggested the actual scale was far larger than anyone had estimated, but that was 2020, and no comparable data point has emerged since.

The National Migration Survey of 2026-27 — announced by the Ministry of Statistics and Programme Implementation as a year-long study beginning in July 2026 — will be the first comprehensive national migration study in nearly two decades. It may provide answers to some of these questions. Or it may reveal, as COVID-19 did, that the reality is even larger than the estimates.

Until then, the best summary of what we know is this: hundreds of thousands of people leave western Odisha every year through a system that meets the legal definition of bonded labor, travel to work sites where conditions meet the international definition of modern slavery, and generate the physical infrastructure of India’s fastest-growing states while their own state remains among its poorest. The system has persisted for at least fifty years. It has survived multiple governments, multiple laws, multiple rescue operations, and multiple election cycles. It will persist until the conditions that created it are addressed — not the sardars, not the kilns, not the lack of enforcement, but the foundational deficit of livelihood that makes the sardar’s advance the best offer on the table.

The dadan road is paved with rational decisions made under irrational constraints. Every family that walks it is choosing the least bad option. The question for Odisha is not why they choose it. The question is why, after fifty years of knowing the answer, it has not built a better option.