English only · Odia translation in progress

Chapter 8: The Return Ticket

On May 1, 2020, the first Shramik Special train from Surat pulled into Odisha carrying migrant workers home. By July 7, 358,401 people had returned by train alone. Another 185,504 had come by bus, private vehicle, or on foot. The state government had prepared 2.27 lakh quarantine beds across 7,200 isolation facilities — schools, community centers, government buildings converted into Temporary Medical Centres in every Gram Panchayat. Each returnee underwent fourteen days of mandatory quarantine and received a Rs 2,000 incentive upon completion. Almost ninety percent of Odisha’s 1,189 COVID cases by late May 2020 were attributed to returning migrants. Cases jumped from 162 to 611 in eleven days after the returnees began arriving.

For a few weeks in mid-2020, something that migration researchers and state governments had discussed for decades appeared to be happening by accident. Mass return. The villages that had been emptied every November were suddenly full in May. Men who had spent twenty years operating powerlooms in Surat were sitting on their verandas in Ganjam. Families that had scattered across three states were together under one roof for the first time in years. “Almost everyone who had returned said they didn’t want to go back to the cities,” reported Mongabay from coastal Odisha.

The desire was real. The economics were not.

Within five months, the workers began trickling back to Surat. “After almost five months of unemployment, workers have now started to travel back,” reported The Wire. The powerloom industry that had ground to a halt without its Odia workforce was reopening. The brick kiln sardars were visiting villages again by October, distributing advances for the next season. By the time the 2020-21 dadan cycle resumed, the great natural experiment in return migration was functionally over.

COVID-19 did not test whether Odia migrants wanted to come home. It proved that they did. What it tested — and what it conclusively answered — was whether home could sustain them. The answer was no.

This chapter is about what it would take to change that answer.

What the COVID Returnees Found

The gap between “home” as an emotional concept and “home” as an economic reality had never been measured so precisely.

A powerloom worker in Surat earns Rs 20,000-25,000 per month, paid at Rs 1.10-1.50 per meter of fabric on piece rate, working twelve-hour shifts. The conditions are terrible — cramped rooms shared with sixty other workers, no safety protections, no health insurance, the constant noise of looms damaging their hearing. But the money is real. After deducting Rs 2,300-2,500 for mess and accommodation, a worker can send Rs 10,000-15,000 home monthly. That money has built pucca houses across Ganjam, put children through school, repaid moneylenders, and funded small enterprises.

What did these workers find when they returned?

MGNREGA — the Mahatma Gandhi National Rural Employment Guarantee Act — was the primary employment option. In 2020, the wage rate was Rs 207 per day. Even at full utilization of the guaranteed 100 days, that yields Rs 20,700 for the year — roughly what a Surat worker earns in a single month. In practice, the average employment generated per household in Odisha was 53 days in 2023-24, down from 55.51 the previous year. At Rs 237 per day (the 2025 rate), that translates to roughly Rs 12,560 in actual annual earnings from the program.

The arithmetic is devastating. A family that migrates to Surat can earn Rs 2.4-3 lakh annually. A family that stays and works MGNREGA earns Rs 12,000-25,000. The gap is not ten or twenty percent. It is tenfold.

Agricultural land could not absorb them either. In western Odisha’s migration-prone districts — Balangir, Kalahandi, Nuapada — the structural problems that drove people out in the first place had not changed during their absence. In Balangir, approximately three percent of agricultural land has irrigation. Ninety-three percent of Odisha’s farming community are small and marginal farmers operating fragmented holdings. Cropping intensity was 117 percent against a national average of 156 percent. A single monsoon-dependent crop, grown on rain-fed land too small to generate surplus — this was what “home” offered in economic terms.

The small towns had no service economy to speak of. Berhampur, the largest city in Ganjam, the district that sends 700,000 workers to Surat, had limited formal sector employment. Balangir town, Kantabanji, Khariar — these are the towns closest to the dadan migration epicenter. They function as transit points for labor heading elsewhere, not as destinations for labor staying put.

Some returnees tried to make it work. Brothers Ananta and Sumanta returned from Pune’s hospitality industry and started a restaurant called Jihoba Taaza Tawa, earning Rs 45,000 in monthly profit. Others used savings to establish internet centers, rice mills, and auto-rickshaw services. Motorcycles and vehicles increased in villages from returnee capital. But these were individual stories of resourcefulness, not a pattern that could absorb hundreds of thousands of workers. The structural conditions that push migration — poor irrigation, absent industry, low wages — remained untouched.

The question was never about desire. Ninety-five point eight percent of migrant households surveyed in Surada block (Ganjam) reported they could not have escaped poverty without migration income. In Baliguda block (Kandhamal), 62.9 percent said the same. When nearly everyone agrees that they cannot survive without leaving, the problem is not the leaving. The problem is the staying.

The Government Response: The Current Toolkit

After twenty-four years of BJD rule during which migration largely went unaddressed as a policy priority, the change of government in 2024 brought migration into political focus. “Dadan Khati” — the distress of laborers forced to leave the state — became a campaign weapon. Naveen Patnaik himself contested from Kantabanji, the epicenter of dadan labor migration, as his second seat. He lost to BJP’s Laxman Bagh by 16,344 votes. The new BJP government under Mohan Charan Majhi has treated migration as a stated priority. The question is whether the tools match the scale of the problem.

The Distress Migration Task Force (October 2024). A twenty-member body led by Deputy Chief Minister K.V. Singh Deo, with Deputy CM Pravati Parida, Panchayati Raj Minister Rabinarayan Naik, Chief Secretary Manoj Ahuja, and Development Commissioner Anu Garg among its members. Its mandate: assess migration dynamics, link welfare schemes to migrant families, improve skills and credit access, coordinate with destination states. It has launched awareness campaigns in thirty migration-prone blocks and conducted official visits to destination states for rehabilitation planning.

Assessment: The task force has high-level political commitment, which is new. But task forces are coordination mechanisms, not implementation bodies. The previous government also identified migration-prone districts and created action plans. The Odisha State Migration Profile Report, prepared with ILO collaboration, dates to 2014. The diagnoses have been available for over a decade. What has been missing is execution — irrigation projects completed, factories built, workers actually earning wages at competitive rates. A task force that produces another set of recommendations without execution infrastructure is, as NewsClick described the Balangir initiative, “a gold filling to hide decay.”

MGNREGA expansion to 300 days in 30 migration-prone blocks. Starting in July 2022, Odisha began providing 200 additional days of work beyond the standard 100-day guarantee in twenty migration-prone blocks of Bargarh, Bolangir, Kalahandi, and Nuapada. In November 2023, this was extended to ten tribal blocks in Gajapati, Kandhamal, Koraput, Nabarangpur, and Rayagada. By 2025, the total coverage reached thirty blocks in nine districts, with up to 300 guaranteed days. The state adds a supplementary Rs 115 per day on top of the central MGNREGA rate of Rs 237, bringing the total to Rs 352 per day. The program has been extended through 2028-29, benefiting 5.57 lakh families.

Assessment: This is the most substantive intervention. At Rs 352 per day for 300 days, a worker could theoretically earn Rs 1,05,600 annually — still less than half of Surat earnings, but a meaningful improvement over the baseline. The problem is delivery. Average actual employment generated per household remains around 53 days, far below even the standard 100-day guarantee. If the government cannot deliver 100 days reliably, promising 300 is aspiration, not implementation. Academic research (Springer, 2023) found that MGNREGA and convergence schemes “appear to have been effective in arresting distress migration, thanks to the rise in household incomes.” But the sharp rise in tracked dadan migration — from 70,142 in 2024 to 94,106 in 2025, a 34 percent increase — suggests the impact remains insufficient against the pull of significantly higher destination wages.

The Shramik Sathi App. A mobile application for migrant worker registration and tracking, designed to create a digital record of who leaves, where they go, and what support they can access.

Assessment: Digital registration solves a real problem — the government has historically been unable to even count its migrants accurately. But the app requires smartphones, internet access, and awareness among a population that is overwhelmingly poor, often illiterate, and recruited through informal networks designed to avoid official scrutiny. The formal registration system under the Inter-State Migrant Workmen Act has been described as “a dead letter” for decades. Whether a new digital version of the same intent will fare better depends entirely on whether sardars and kiln owners cooperate — which they have no incentive to do.

Odisha TEX 2025. The state’s textile industry initiative produced thirty-three MoUs worth Rs 7,808 crore with projected creation of 53,300 direct jobs. Worker subsidies of Rs 6,000 per month for male workers and Rs 7,000 for female workers. Hindalco committed Rs 200 crore for 2,400 jobs. The target is 100,000 jobs by 2030.

Assessment: This is the intervention most directly targeted at migration reversal. If Ganjam workers operate powerlooms in Surat, building a textile industry in Odisha is the obvious structural response. The numbers on paper are significant. But the distance between an MoU and an operational factory employing workers is vast. Odisha has a history of investment announcements that do not materialize — POSCO’s Rs 52,000 crore steel plant, ArcelorMittal’s proposed investment, the FDI gap (Rs 67,000 crore in announced commitments versus Rs 39 crore in actual FDI in 2024-25, per some estimates). The test for Odisha TEX is not how much was signed but how many of those 53,300 jobs exist and pay competitive wages three years from now. That data does not yet exist because the programs are too new.

OSDA skill development programs. The Odisha Skill Development Authority, established in 2016 with a goal of skilling 6.7 lakh youth in two years, runs programs under PMKVY, CMEGP, and its Placement Linked Training Program. Supported by the Asian Development Bank through the Odisha Skill Development Project. District-level Skill Gap Analysis has been completed for all thirty districts.

Assessment: Skill development is necessary but not sufficient. Training workers for jobs that do not exist locally does not reduce migration — it may accelerate it by making workers more employable elsewhere. The Swakalpa program, a two-year self-employment initiative funded by ADB, demonstrated that grassroots entrepreneurship can work — seventy percent of supported businesses were women-led, far exceeding the twenty-five percent target. But these are pilot-scale programs. Scaling them to absorb hundreds of thousands of returning workers requires a level of institutional capacity and sustained funding that has not been demonstrated.

Mobile Migrant Resource Centres and MiRC network. A toll-free helpline (1800-345-7885) and a network of eleven NGOs tracking migration from western and coastal Odisha, following up rehabilitation of released bonded laborers in seven districts.

Assessment: These are support mechanisms for a crisis that continues, not solutions to the crisis itself. They help migrants navigate a system that should not require navigation. The MiRC network provides ground-truth data that government statistics miss, and the helpline gives workers access to information about rights and services. Both are valuable. Neither addresses why people leave.

What Returnees Actually Do

The success stories are real. They are also rare.

Sisir Gouda worked in textile mills in Surat and Mumbai for thirty-two years. In 2020, he returned to Balakrushnapur village in Ganjam and established Matexmate Textile Private Limited. With two partners, he invested Rs 2 crore — self-funded, from three decades of savings — to install nine weaving machines on 3,000 square feet of land. The capacity can expand to fifty machines, accommodating more than two hundred textile workers. Sisir is seeking state government financial support to scale up.

This is the story that policymakers love to cite. A migrant returns with skills, capital, and industry knowledge, and creates an enterprise that could employ other migrants. It is an important proof of concept. But consider the conditions it required: thirty-two years of savings, specific transferable skills in an industry that can function at small scale, entrepreneurial drive, and two partners willing to co-invest. These are not conditions that the typical dadan worker from Balangir or even the typical powerloom operator from Ganjam can replicate.

Shramavahini — the Odisha chapter of the Released Bonded Labourers Association — represents something different. With over a thousand registered survivor members, it is the organized voice of those who have endured the worst of the migration system. Members engage with government officials on rescue and rehabilitation, conduct awareness campaigns through media and social media, and work to empower survivors to access welfare programs. Shramavahini matters because it is not a program imposed from above but an organization built by the people most affected. It addresses the immediate crisis of exploitation, not the root cause of departure.

The Swakalpa model demonstrated that with mentorship and financing, grassroots entrepreneurship can work. But it operated at pilot scale. The overwhelming pattern for returnees is simpler and less encouraging: most either go back to migration within months, or subsist on MGNREGA and marginal agriculture. The COVID natural experiment proved this at scale. The desire to stay is real. The economic scaffolding to support staying is absent.

The Wage Problem: The Core Structural Barrier

Everything else is subordinate to this.

A worker in Surat earns Rs 20,000-25,000 per month. MGNREGA pays Rs 237 per day — about Rs 5,925 per month if a worker gets twenty-five days of employment, which most do not. Even with the supplementary Rs 115 per day in high-migration blocks, the total reaches Rs 352 per day or roughly Rs 8,800 per month at twenty-five days. A factory job in Odisha — if one existed — might pay Rs 10,000-15,000 per month for unskilled or semi-skilled work, based on Odisha’s current minimum wage of Rs 452 per day for unskilled workers.

The wage gap, in other words, is fifty to one hundred percent. A Surat worker earns two to four times what the same person could earn in Odisha through any available formal or informal channel.

This gap is not a market failure that can be corrected by a subsidy or a training program. It reflects the fundamental difference between a state with a dense industrial ecosystem and a state without one. Surat’s powerloom industry produces ninety percent of India’s polyester. The industry exists because of decades of accumulated capital, supply chain infrastructure, market access, and labor pool depth. A worker there earns what a worker earns because the value chain around them is deep enough to sustain those wages. Odisha’s agricultural and informal economy cannot generate equivalent value per worker.

The realistic threshold for reversing migration is not wage parity. It is perhaps seventy to eighty percent of destination wages, combined with the non-monetary advantages of being home — lower living costs, family proximity, social capital, cultural rootedness. If an Odia worker could earn Rs 15,000-18,000 per month at home while living with family, in a house already owned, without paying Surat’s mess and accommodation fees, the economic calculation might shift. The total package — wages plus lower costs plus family life — could become competitive.

What industries could generate these wages in Odisha?

Textiles are the most obvious candidate, and Odisha TEX 2025 targets this directly. The logic is straightforward: if 700,000 Odias already know how to operate powerlooms, and the raw material (polyester yarn) can be shipped to Odisha, then building a textile manufacturing base in Ganjam and surrounding districts could reabsorb a significant portion of the Surat workforce. The worker subsidy of Rs 6,000-7,000 per month is designed to bridge the initial wage gap while the industry scales. But textile manufacturing requires more than looms — it requires power supply reliability, water access, logistics infrastructure, market connectivity, and the cluster effects that make Surat competitive. Building an industrial cluster takes a decade, not a budget cycle.

Food processing could absorb agricultural surplus and create off-farm employment. Odisha produces significant volumes of rice, turmeric, cashew, and seafood. Processing these locally — rather than exporting raw commodities — would add value and create jobs at wages above agricultural labor rates. The food processing industry typically pays Rs 8,000-15,000 per month for semi-skilled workers, which approaches the lower end of the target range.

Electronics assembly and garment manufacturing are labor-intensive industries that have transformed employment in Tamil Nadu, Karnataka, and Andhra Pradesh. They require industrial parks, power, transport links, and a basic skills pipeline — all achievable but not yet built at the required scale in western Odisha’s migration-source districts.

The critical point about industrial job creation is geography. Building factories in Bhubaneswar or even the Kalinganagar industrial corridor does not directly address migration from Balangir, Ganjam, or Nuapada. These districts are between 200 and 400 kilometers from the existing industrial clusters. The model that would actually reduce migration is industrial clusters located in or near the migration-source districts themselves — textile parks in Ganjam, food processing in Kalahandi, light manufacturing in Balangir. This is harder than concentrating investment in existing infrastructure corridors, but it is the only approach that directly addresses the geographic mismatch between where jobs are and where migrants come from.

I should be honest about confidence levels here. Whether Odisha can build competitive industrial clusters in its currently underdeveloped western and southern districts within a decade is uncertain. The conditions that make industrial clusters work — reliable power, water, roads, a pool of trained workers, proximity to markets — take time and sustained capital investment to create. I would estimate, with roughly forty percent confidence, that Odisha TEX and related programs will deliver at least 25,000 of their targeted 53,300 textile jobs by 2030. The remaining gap will need to be filled by complementary sectors or will remain unfilled.

The Ecosystem Problem: More Than Wages

For the powerloom worker, wages are the primary variable. But for the IT professional in Bangalore, the doctor in Hyderabad, the engineer in Pune — the six lakh Odias in Bangalore alone — the calculation is more complex.

The Bangalore IT professional will not return for a twenty percent pay cut if the schools in Bhubaneswar are worse, the hospitals are unreliable, the power goes out daily, and the streets flood in light rain. These are not hypothetical complaints. “There are powercuts everyday, water logging even for slightest rain, and water shortage with dependency on water tankers,” as one Reddit discussion about Bhubaneswar noted. Healthcare access in rural Odisha is poor — fifty-four percent of the population seeks private healthcare because public facilities have “inconvenient locations, lower stocks, lack of respect from providers, poorer infrastructure.” There is a documented gap of 203,496 hospital beds and 4,432 diagnostic centers.

The ecosystem problem compounds the wage problem. Even if a Bhubaneswar IT company offered a salary at ninety percent of Bangalore levels, the professional considering the move would also calculate: can my children get quality education here? Will my spouse find comparable employment? Can I access specialist medical care when needed? Is the city connected enough — flights, internet, cultural life — that I do not feel isolated from the professional mainstream?

This ecosystem gap is why Bhubaneswar, despite hosting over 200 IT firms including TCS, Infosys, Wipro, Tech Mahindra, Mindtree, and Capgemini, and despite operating Infocity SEZ (145 acres) and Info Valley SEZ (262 acres), has not become a serious alternative to Bangalore for most Odia IT professionals. The companies are present, but the ecosystem — the density of specialized roles, the startup culture, the professional network effects, the quality of life infrastructure — lags by a decade or more.

This matters because the skilled departure, though quieter than dadan migration, may be more structurally damaging. When NIT Rourkela produces 1,274 job offers with an average salary of Rs 14.10 lakh per annum and virtually all high-package placements are in Bangalore, Hyderabad, or Pune, the state is investing in training talent that will build other states’ economies. When 42 IAS posts and 67 IPS posts lie vacant in the Odisha cadre, with 62 officers on central deputation, the administrative capacity to implement any development program is degraded.

The ecosystem problem does not have a single solution. It requires sustained, parallel investment across education (schools that professionals would choose for their children), healthcare (hospitals that do not require a flight to Bangalore for serious treatment), urban infrastructure (reliable water, power, drainage, public transport), and quality of life (cultural venues, restaurants, parks, connectivity). These are the unsexy investments that do not produce MoU headlines but determine whether skilled people choose to live in a place.

What Other States and Countries Did

The question of whether return migration can be engineered at scale is not theoretical. Several states and countries have attempted it, with varying degrees of success.

Kerala’s NORKA-ROOTS model. Kerala — which sends approximately 2.12 million workers abroad, primarily to the Gulf — created the Department of Non-Resident Keralites’ Affairs (NORKA) and its operational wing NORKA-ROOTS to provide institutional support for returnees. The system includes a Pravasi Welfare Board, insurance schemes for overseas workers, pension programs, skill verification, and mechanisms for channeling Gulf remittances into productive investment through the cooperative sector. Kerala conducts systematic migration surveys — five rounds since 1998 — that give the state government actual data on migration flows, skill profiles, and returnee needs. Remittances constitute 22-28 percent of Kerala’s state GDP.

What is transferable: the institutional architecture. Odisha has nothing comparable to NORKA — no dedicated department for its millions of migrants, no systematic survey infrastructure (the 2023 OMS survey by IIT Hyderabad was a first step), no welfare board, no insurance mechanism. The NRO Cell in the Chief Resident Commissioner’s office in Delhi is a liaison, not a service delivery institution. Kerala demonstrates that treating migration as a governance priority rather than a political embarrassment changes what is possible.

What is not directly transferable: Kerala’s migration is predominantly international and high-wage. Its returnees bring back significant capital. Odisha’s migration is predominantly domestic and low-wage. A dadan worker returning from a brick kiln with depleted savings is in a fundamentally different position from a Kerala construction worker returning from Qatar with five years of Gulf earnings.

China’s interior development strategy. Beginning in the late 1990s, China launched the “Develop the West” campaign, investing massively in infrastructure, industry, and urbanization in its interior provinces to slow the coastal migration that was depopulating rural areas. The Great Western Development Strategy included railways (the Chengdu-Kunming line), highways, power generation, and industrial relocation incentives. The scale was enormous — hundreds of billions of dollars over two decades. The result: while coastal cities remain richer, the interior provinces have seen significant industrialization and urbanization, and the rate of net outward migration from western China has slowed substantially.

What is transferable: the principle that reducing migration requires building economic activity at the source, not just managing the flow. The idea that industrial clusters can be deliberately created in underdeveloped regions through sustained infrastructure investment is proven.

What is not transferable: China’s scale of state investment and its capacity for top-down implementation are not replicable in India’s federal democratic system. An Indian state government cannot command factories to relocate to Balangir the way the Chinese government can direct state enterprises to Chengdu. The principle applies; the mechanism does not.

Mexico’s Programa 3x1. Mexico created a matching-funds program where the federal, state, and municipal governments each match every dollar that migrant organizations invest in community projects. A diaspora group that raises $100,000 for a water system in their home village triggers $300,000 in government matching funds. The program channels remittances into community infrastructure — roads, water, electricity, schools — rather than solely into private consumption.

What is transferable: the idea of leveraging remittances for community development. Ganjam alone receives an estimated Rs 120-124 crore per month in remittances — roughly Rs 1,440-1,488 crore annually from one district. This money currently goes primarily to housing construction, debt repayment, education, and consumption. A matching program could channel even a small fraction of these flows into productive community infrastructure — irrigation, market yards, cold storage, food processing facilities — that would improve the economic viability of staying.

What is not transferable: Mexico’s program works through organized migrant associations with institutional capacity to plan and execute community projects. Odisha’s Surat migrants, despite numbering 700,000-800,000, have no formal community organization comparable to Mexico’s hometown associations. Over 500 Odia community groups operate in Surat, but these are fragmented along caste and village lines and function primarily as cultural organizations, not development entities.

The honest synthesis of these comparisons is that no model translates directly, but each proves that specific elements of the return-migration problem have been solved elsewhere. Institutional support for migrants (Kerala), infrastructure investment at the source (China), and leveraging remittances for community development (Mexico) are all proven approaches. Odisha uses none of them at meaningful scale.

The Honest Assessment

Migration from Odisha will not reverse in the short term. This is not pessimism. It is arithmetic.

The structural conditions that drive migration — poor irrigation, absent industry, low wages, weak infrastructure, inadequate urban development in tier-2 cities — took decades to create. The freight equalization policy that prevented Odisha from building a steel industry lasted from 1952 to 1993 — forty-one years. The underinvestment in irrigation that makes western Odisha agriculture unviable has persisted since independence. The gap between mineral extraction and mineral processing that could have created downstream jobs has been a consistent policy failure across multiple governments, central and state. These are not problems that yield to a task force or a textile MoU.

The realistic goal is not zero migration. Some migration is rational and beneficial — the NIT Rourkela graduate who takes a Google placement in Bangalore is exercising choice, not fleeing desperation. The realistic goal is making migration a choice rather than a compulsion. The test: when a worker in Ganjam can choose between a textile job at home paying Rs 15,000-18,000 per month and a Surat loom paying Rs 20,000-25,000, and the total life quality — family proximity, lower costs, cultural belonging — makes the home option genuinely competitive, the system is moving in the right direction. When a family in Balangir can grow two crops a year because irrigation has reached their land, and the combination of farm income and MGNREGA work generates Rs 10,000-12,000 per month, the sardars’ advance loses its power.

I should state what would have to be true for me to be wrong — for the optimistic scenario to unfold faster than I expect. It would require that Odisha TEX and related industrial programs deliver jobs at scale and on time, something the state has never demonstrated. It would require that the MGNREGA expansion to 300 days actually delivers 300 days of employment, not the current average of 53. It would require that irrigation investment in western Odisha accelerate dramatically — and the track record of irrigation project completion in India, let alone Odisha, gives little basis for confidence. It would require that the current BJP government sustains its focus on migration beyond the next election cycle, treating it as a generational project rather than a campaign talking point.

If all these conditions converge, meaningful reduction in distress migration is possible within a decade. I estimate the probability of that convergence at roughly twenty to twenty-five percent. More likely — perhaps sixty percent probability — is a scenario where partial progress occurs: some textile jobs materialize but fewer than announced, MGNREGA delivery improves but remains below targets, irrigation expands incrementally, and migration continues but at a somewhat reduced rate in the targeted blocks while continuing to grow in districts not covered by the interventions. The remaining fifteen to twenty percent probability covers a scenario where execution fails entirely, political attention shifts, and migration rates continue to increase as they did between 2024 and 2025.

What Would Actually Work: Structural Conditions

If sentiment is not strategy, what is? Four structural shifts would have to occur simultaneously. None is sufficient alone.

First: labor-intensive manufacturing in migration-source districts. Not in Bhubaneswar. Not in Kalinganagar. In Ganjam, Balangir, Nuapada, Kalahandi — where the people actually leave from.

Textiles are the obvious starting point because of the existing skill base in Surat. An industrial cluster in Ganjam that employs even 20,000 workers at Rs 12,000-15,000 per month would create a viable alternative to Surat for a fraction of the 700,000 Ganjam workers there. Food processing is the second candidate — Odisha produces turmeric, cashew, rice, and seafood in volumes that justify processing facilities. Electronics assembly and garment manufacturing are the third tier — proven job creators in Tamil Nadu and Karnataka that could work in Odisha with the right infrastructure.

The model: dedicated industrial parks in or near the high-migration districts, with guaranteed power supply, road connectivity, water access, and a five-year tax incentive comparable to what other states offer. The worker subsidy from Odisha TEX (Rs 6,000-7,000 per month) is a good bridge mechanism, but the subsidy eventually has to be replaced by self-sustaining wage economics. If the industry cannot pay competitive wages without permanent government subsidy, it is not a solution — it is a transfer payment.

Second: agricultural transformation in western Odisha. The single highest-impact intervention for dadan migration is irrigation. In a district where three percent of agricultural land is irrigated, expanding that to thirty percent would transform the agricultural economy — enabling a second (rabi) crop, supporting crop diversification beyond rain-fed paddy, and making food processing viable by creating a reliable supply of raw material.

The numbers make the case. Kharif irrigated area in Odisha is 30.89 lakh hectares; rabi irrigated area is only 15.01 lakh hectares — less than half. The gap between kharif and rabi irrigation is the gap between a single-crop economy and a dual-crop economy. Closing it in the migration-prone districts would directly increase farm incomes and reduce the push factor that sends families to brick kilns.

Irrigation projects in India are infamous for cost overruns and decades-long delays. Major projects funded during the Vajpayee era — Lower Suktel, Telengi, Mahendratanaya — remain incomplete. The track record is poor. But smaller-scale solutions — check dams, farm ponds, micro-irrigation, groundwater development — can be implemented at block level without the mega-project timelines. The Odisha Economic Survey 2025-26 reports total irrigated area reaching 74.2 lakh hectares, suggesting improvement, but the critical question is whether that improvement is reaching the districts that need it most.

Third: urban development in tier-2 cities. Berhampur, Sambalpur, Rourkela, Balangir — these towns need to become livable urban centers with services and economic opportunities. This means reliable municipal services (water, sanitation, electricity, transport), educational institutions that retain rather than lose talent, healthcare facilities that do not require a flight to Bhubaneswar for specialist care, and commercial ecosystems that support small and medium enterprises.

Currently, these towns function as administrative centers and transit points, not as economic engines. Transforming them requires sustained municipal investment over at least a decade — the kind of patient capital that does not produce headlines but creates the conditions for private investment and employment growth.

Fourth: connectivity. Roads, rail, and internet are the basic infrastructure that makes everything else possible. The Talcher-Bimlagarh railway line, approved over seventy years ago, remains incomplete — a fact that symbolizes the connectivity deficit more eloquently than any statistic. Road connectivity in western Odisha is poor; only 28 percent of Odisha’s roads are surfaced. Internet penetration, while growing, remains inadequate for digital economy participation in rural areas.

These four shifts — manufacturing, agriculture, urbanization, connectivity — are not novel recommendations. They have appeared in every development plan for Odisha for decades. The question is not what needs to be done. The question is whether the political will and institutional capacity exist to execute them at sufficient scale for sufficient duration. History suggests that they do not, but the 2024 political transition and the federal context of competitive federalism create conditions that are at least different from the past, even if not demonstrably better.

The Political Economy of Migration

“Dadan Khati” entered Odisha’s political vocabulary as an electoral weapon in 2024. The BJP used it to indict twenty-four years of BJD governance — asking why, after a quarter century in power, the state still could not keep its workers. The argument was effective. Naveen Patnaik’s defeat in Kantabanji, the symbolic heart of dadan migration, was read as a verdict on the failure to address distress migration.

The new government has responded with announcements: the task force, Odisha TEX, MGNREGA expansion, the Shramik Sathi app. These are real programs with real budgets. But the pattern that Odisha TEX must break is the pattern that has defined Odisha’s development for decades: the announcement economy.

Consider the arithmetic. Rs 7,808 crore in MoUs for 53,300 textile jobs sounds significant. But Odisha has seen this before. POSCO signed an agreement for Rs 52,000 crore in 2005 and never built a plant. ArcelorMittal proposed a similarly massive investment and withdrew. The gap between committed FDI and actual FDI is one of the most consistent features of Odisha’s economic history. In some estimates, announced investment commitments exceed actual foreign direct investment by a factor of more than a thousand.

The difference between a political acknowledgment and a structural solution is the difference between diagnosis and treatment. The BJP government has correctly diagnosed the problem — publicly, loudly, and with electoral consequences. Whether it can sustain the institutional focus, capital investment, and execution discipline to actually build 53,300 jobs (or even a fraction of that number) over the next five to ten years will determine whether the diagnosis was the beginning of a cure or another iteration of the announcement cycle.

The political incentive structure is not encouraging. Elections occur every five years. Industrial clusters take a decade to mature. The political reward for announcing an MoU is immediate and photogenic. The political reward for completing an irrigation canal that comes online eight years later accrues to whoever happens to be in power when the water flows — which may be a different party. This temporal mismatch between political cycles and development timelines is not unique to Odisha. But it is particularly acute for migration reversal, which requires the kind of patient, sustained investment that democratic politics systematically undervalues.

Furthermore, approximately 200,000 voters in western Odisha are effectively disenfranchised by migration. They are registered in their home constituencies but absent during elections. Berhampur, Ganjam’s largest city, recorded only 63 percent turnout despite 1.6 million eligible voters — the lowest, attributed to migrant absence. Workers who cannot vote from their work locations have no mechanism to hold any government accountable for the conditions that forced them to leave. The people most affected by migration policy are least able to influence it through electoral means.

What the Sadhabas Would Recognize

There is a structural parallel between ancient Kalinga’s maritime economy and modern Odisha’s migration economy that is worth drawing, not for sentimental reasons but for analytical ones.

The sadhabas — the Kalingan maritime traders who sailed to Southeast Asia — exported value-added goods. Textiles, metalwork, ivory, semi-precious stones, and most importantly, knowledge — Buddhism, temple architecture, scripts, dance forms, agricultural techniques. They returned with profit, with new goods, with cultural exchange. The trade was balanced. Both sides created and captured value. Kalinga’s economy grew from the exchange because what it exported was worth more than its raw-material cost. The value was in the processing, the skill, the knowledge embedded in the goods.

Modern Odisha exports two things in volume: raw minerals and raw labor. Iron ore leaves the state at four thousand rupees per tonne and returns as steel at seventy thousand. Bauxite leaves and returns as aluminum. Coal leaves and returns as electricity. And people leave — 700,000 to Surat alone — and their labor returns as remittances that represent a fraction of the value their work creates in Gujarat’s textile economy.

The pattern is the same: extraction without processing. Whether the raw material is iron ore or human labor, Odisha provides the input and other states capture the value-add. The sadhabas would recognize the problem immediately. They would see a state that has forgotten how to process its own resources — mineral and human — into products worth more than the raw input.

The return voyage — whether across the Bay of Bengal a thousand years ago or from Surat today — requires the same structural condition: creating value at home that is worth more than what can be earned by sending raw inputs elsewhere. For the ancient trader, that meant textiles fine enough to be prized in Java. For the modern state, it means an economy productive enough to employ its own people at wages that justify staying.

The Return Ticket

The return ticket is not a train ticket from Surat to Berhampur. It is not a government scheme or a task force report or an MoU signed at an investment summit. The return ticket is an economy that gives people a reason to stay.

Building that economy is the work of a generation, not a budget cycle. It requires irrigation that reaches the three-percent-irrigated fields of Balangir. It requires textile mills in Ganjam that can pay seventy percent of Surat wages while offering the non-monetary dividend of home. It requires tier-2 cities that function as livable urban centers rather than transit points. It requires schools good enough and hospitals reliable enough that a professional returning from Bangalore does not feel they are sacrificing their family’s future. It requires roads and rail connections that allow industrial activity to take root in places that currently function only as labor-export zones.

None of this is impossible. Kerala built institutional support for its migrants. China invested in its interior. Mexico leveraged remittances for community development. Tamil Nadu and Karnataka built industrial ecosystems that attract workers from Odisha. The knowledge of how to build economies exists. What has been missing in Odisha is not the knowledge but the execution — sustained, patient, compounding investment over decades rather than announcement-cycle bursts.

The question is whether the political will and institutional capacity exist to sustain the effort for that long. The 2024 political transition created a window. “Dadan Khati” is now a live political issue. The central and state governments are both making commitments. The National Migration Survey 2026-27 will provide the first comprehensive migration data in nearly two decades, creating the baseline against which progress can be measured.

But windows close. Political attention shifts. The next crisis, the next election, the next set of priorities crowds out the patient work of building irrigation canals and industrial parks and municipal water systems. The history of Odisha’s development is a history of windows that opened and closed without the sustained institutional effort needed to walk through them.

The COVID experiment delivered its verdict. Given the choice, Odia migrants want to come home. The question was never about desire. The question was always about whether home would be ready for them. Five years after the pandemic, the answer is still being constructed — in textile MoUs that may or may not produce factories, in MGNREGA expansions that may or may not deliver promised workdays, in irrigation projects that may or may not reach the fields that need them, in an economy that is growing at 7.9 percent but not yet generating the specific kind of jobs, in the specific places, at the specific wages that would make the return ticket worth buying.

The sadhabas’ boita carried them across the Bay of Bengal because there was value on the other side worth the journey. The return journey happened because there was value at home worth coming back to. When Odisha rebuilds that value — not in ceremonial memory but in functioning factories, irrigated fields, and livable towns — the return tickets will sell themselves. Until then, the trains from Berhampur to Surat will continue to run full, and every Kartik Purnima, the small boats will float on the Mahanadi, commemorating a departure that has not yet learned how to become an arrival.

Sources

COVID-19 Migration and Return Data

Government Programs and Policy

Wage and Employment Data

Migration Scale and Tracking

Returnee Case Studies

Agricultural and Infrastructure Data

Surat Corridor

Political Economy

Comparative Models

Skill Development

Research Surveys

IT and Skilled Migration

Source Research

The raw research that informs this series.