English only · Odia translation in progress

Chapter 5: The Labor Gap — The Workers Who Leave and the Skills That Don’t Exist

In the 2023-24 placement season, NIT Rourkela’s Department of Metallurgical and Materials Engineering — one of India’s oldest and most respected programs in the discipline — placed the overwhelming majority of its roughly 120 graduates in companies headquartered outside Odisha. Tata Steel Jamshedpur, JSW Steel Vijayanagar, Hindalco Renukoot, SAIL Bokaro, the Pune auto component cluster, and the Bangalore tech corridor absorbed them. A smaller number went abroad for graduate studies in materials science. The number who stayed to work in Odisha’s own minerals and metals industry could, in most years, be counted on two hands.

This is not a failure of the institution. NIT Rourkela’s metallurgical engineering department has been producing world-class metallurgists since 1961, when it was established to supply technical manpower for the Rourkela Steel Plant built with German collaboration. The irony is structural and precise: the department trains metallurgists who go to Jamshedpur to process Odisha’s own iron ore into steel. The ore travels south-to-north by rail. The engineers travel the same route by placement offer letter. The raw material and the human capital both leave the state, and both end up adding value to someone else’s economy.

This chapter is about the workforce that would need to exist for Odisha to process its own minerals — and the gap between what exists and what is needed. The gap is not primarily one of population. Odisha has 46 million people. It is a gap of skills, wages, and industrial structure. The labor is not missing. It is in the wrong places, doing the wrong things, for the wrong states.

What Mining Employment Actually Looks Like

Start with what exists. Odisha’s mining sector — across coal, iron ore, chromite, bauxite, manganese, limestone, and quarrying — employs somewhere between 150,000 and 200,000 people in formal and semi-formal roles. The number is imprecise because the line between formal and informal in Indian mining is deliberately blurred.

The largest single employer is Mahanadi Coalfields Limited (MCL), the Coal India subsidiary that operates the Talcher and Ib Valley coalfields. MCL employs approximately 20,000 permanent workers. But that number tells less than half the story. The mine also relies on a large contract workforce — workers hired through third-party contractors who provide none of the benefits (housing, medical, pension, job security) that permanent MCL employees receive. The contract workers do the same physical work, in the same pits, breathing the same coal dust. They earn a fraction of the permanent workers’ wages. This dual workforce structure — a thin layer of well-paid permanent employees atop a vast base of poorly-paid contract workers — is not unique to MCL. It is the standard operating model across Indian mining.

NALCO, the state’s other major public sector mineral enterprise, employs roughly 7,000-8,000 people across its bauxite mines at Panchpatmali, its alumina refinery at Damanjodi, and its smelter at Angul. These are among the better industrial jobs in Odisha — public sector wages, housing colonies, medical facilities. But 7,000-8,000 jobs across a company that processes one of the world’s largest bauxite deposits is a remarkably small number. NALCO’s operations are highly mechanized. The bauxite mine at Panchpatmali, sitting at 1,200 metres elevation on the Eastern Ghats, uses large earth-moving equipment that requires few operators per tonne extracted.

Private mining companies — the ones that won lease auctions post-2015 — employ even more skewed workforces. In a typical private iron ore mine in Keonjhar, 70 percent or more of the workforce is contract labor. The permanent staff consists of a mining engineer or two, a geologist, an environmental officer (to handle compliance paperwork), a few equipment operators, and a small administrative team. Everyone else — the people who drive the dump trucks, load the conveyors, maintain the roads, run the canteens, handle the security — is contract.

The wage structure tells the story of where value is and is not created in this economy:

Role	Typical monthly income (Rs)
Unskilled mining laborer (daily wage)	10,000-15,000
Skilled machine operator	20,000-30,000
Supervisor / foreman	25,000-40,000
Mining engineer	50,000-100,000
Mine manager / senior geologist	100,000-200,000

The shape of this pyramid is the shape of a mining economy. A vast base of people earning subsistence wages doing physically dangerous work. A thin middle layer of skilled operators. A thinner layer of engineers and managers — many of whom are transferees from other states, posted to Odisha for a tour of duty before moving on to more desirable locations.

A programmer would recognize this as a highly skewed distribution — the kind of data shape that, if it appeared in a codebase’s performance metrics, would trigger an investigation into whether the architecture was fundamentally wrong. The architecture is fundamentally wrong. An economy structured around extraction produces extraction-shaped employment: a few high-skill positions at the top, almost nothing in the middle, and a mass of undifferentiated labor at the bottom.

What Processing Employment Looks Like

Now look at what a processing economy produces in employment terms, and the contrast becomes visceral.

Tata Steel Jamshedpur — the plant that has been processing Odisha’s iron ore since 1907 — employs approximately 35,000 people directly. But direct employment is not the right number to look at. Around the plant, over the past century, an entire industrial ecosystem has grown: equipment manufacturers, refractory suppliers, industrial gas companies, calibration labs, IT service providers, logistics firms, food services, educational institutions, healthcare providers. The estimated indirect employment generated by Tata Steel’s Jamshedpur operations exceeds 150,000 people. Some estimates put the total number of livelihoods dependent on the plant — direct, indirect, and induced — at over 300,000.

Within Odisha, the comparison is instructive:

JSPL Angul — the Jindal Steel and Power complex that is Odisha’s largest private-sector steel operation — employs approximately 10,000 people directly. This is a modern integrated steel plant with a 6 MTPA crude steel capacity, pellet plant, DRI units, and power generation. It is substantial. But it is one plant.

Rourkela Steel Plant (SAIL) — the Nehruvian relic that was Odisha’s first major industrial installation — employs approximately 15,000 people directly, down from a peak of over 25,000 in its heyday. The decline is partly automation, partly the slow attrition of a public-sector unit that has not expanded at the pace of private-sector competitors.

Tata Steel Kalinganagar — the newer plant in Jajpur district, currently at 3 MTPA and expanding to 8 MTPA — employs roughly 5,000-7,000 people directly, with the number growing as expansion phases come online.

Add these up with smaller steel operations (sponge iron plants, EAF-based mills) and Odisha’s total formal steel industry employment is perhaps 40,000-50,000 direct jobs. Compare this with the state’s mining employment of 150,000-200,000 and a superficial observer might conclude that mining employs more people. But the comparison is misleading in a way that matters enormously.

A steel plant creates 3-5 times more jobs per unit of economic output than a mine, at 2-4 times higher average wages. The reason is elementary: processing requires more steps, more skills, and more supporting services than extraction. A mine digs material out of the ground and loads it onto a truck. A steel plant takes that material through iron-making, steel-making, casting, rolling, finishing, quality testing, packaging, and dispatch — each stage requiring operators, technicians, engineers, and support staff.

And then there is the multiplier. For every direct job in a steel plant, approximately 3-4 indirect jobs are created in the surrounding economy — suppliers, service providers, transport operators, food services, education, healthcare. A mine creates roughly 1.5-2 indirect jobs per direct job. The multiplier is lower because a mine’s supply chain is simpler: equipment, fuel, explosives, and labor. A steel plant’s supply chain includes all of those plus refractory materials, alloying elements, industrial gases, specialized instruments, automation systems, and a constellation of technical services.

The wage comparison is where the argument becomes unanswerable:

Role	Location	Monthly income (Rs)
Steel plant technician	Jamshedpur	30,000-50,000
Rolling mill operator	Jamshedpur / Angul	40,000-60,000
Quality control chemist	Jamshedpur / Kalinganagar	50,000-80,000
Maintenance engineer (mechanical)	Angul / Rourkela	60,000-100,000
Process control engineer	Jamshedpur	70,000-120,000

A steel plant technician in Jamshedpur earns Rs 30,000-50,000 per month — two to four times what an unskilled mining laborer earns in Keonjhar, fifty kilometres away. They both work with iron ore. One handles the raw material. The other handles the transformed product. The transformation — and the wage premium it commands — is precisely the value that leaves Odisha when the ore does.

The Skills That Don’t Exist

If Odisha wanted to process all its minerals domestically — a theoretical exercise, but a useful one — what workforce would it need? The answer reveals a skills gap so wide that it resembles less a gap and more a void.

Start with metallurgical engineering capacity — the highest-skill layer of the metals processing workforce.

NIT Rourkela graduates approximately 120 metallurgical and materials science engineers per year. It is one of India’s premier programs. Where do those graduates go? The placement data, while not published in granular geographic detail, tells a consistent story through campus recruiters: Tata Steel Jamshedpur, JSW Steel Vijayanagar (Karnataka), SAIL plants across multiple states, Hindalco facilities in Uttar Pradesh and Maharashtra, auto component companies in Pune and Chennai, and increasingly, IT and consulting firms in Bangalore and Hyderabad. The overwhelming majority leave Odisha. The few who stay typically join JSPL Angul, Tata Steel Kalinganagar, or Rourkela Steel Plant — the small number of processing facilities that actually exist within the state.

IIT Bhubaneswar, established in 2008, is still building capacity. Its School of Minerals, Metallurgical and Materials Engineering produces graduates who, following the gravitational pull of India’s job market, mostly leave the state. The institution is young. Its alumni network is small. Its capacity to anchor a workforce pipeline within Odisha is, at present, minimal.

The irony deserves repetition because it is the kind of structural absurdity that, once seen, cannot be unseen: NIT Rourkela trains metallurgists who go to Jamshedpur to process Odisha’s own iron ore. The state invests in the education — the campus, the faculty, the lab equipment, the subsidized tuition. The state bears the cost. And the return on that investment accrues to Jharkhand, Karnataka, Maharashtra, and Gujarat, where the processing happens.

In software terms, this is a memory leak. Resources are allocated but never freed for the allocating system’s use. The system that produces the resource (Odisha’s educational institutions) does not benefit from the resource’s deployment (the graduates’ productive work). Over time, the leak compounds. Each graduating class that leaves represents not just lost human capital but lost network effects — the senior engineers who would have mentored juniors, the experienced managers who would have founded supplier companies, the accumulated institutional knowledge that makes an industrial cluster self-reinforcing.

Below the engineering layer, the gap widens further.

Industrial Training Institutes (ITIs) — the institutions that produce the fitters, welders, electricians, machinists, and technicians who form the backbone of any manufacturing economy — are a useful proxy for a state’s industrial workforce capacity.

Odisha has approximately 49 government ITIs plus a number of private ones. Tamil Nadu has over 120 government ITIs. Gujarat has over 80. Maharashtra has over 100. Karnataka has approximately 90. The raw numbers are stark, but the real problem is subtler: Odisha’s ITIs produce workers trained for trades that the state’s economy cannot absorb. A welder trained at an ITI in Keonjhar has two choices — take a contract job at a mine (where welding requirements are basic) or migrate to a state where manufacturing exists (where the welding is complex, specialized, and well-paid). The ITI system produces the supply. The absence of manufacturing means there is no demand. The supply migrates.

The skills that do not exist in Odisha at any meaningful scale include:

CNC machine operators — for auto component and precision manufacturing. Odisha has near-zero auto component production, so it has near-zero CNC operators.
Welding technologists — not welders who join two pieces of metal, but specialists who design and program automated welding lines for continuous production. The distinction matters: a welder is a skilled laborer; a welding technologist is an industrial engineer. The former earns Rs 15,000-25,000 per month; the latter earns Rs 50,000-80,000.
Quality assurance engineers — people trained in ISO/TS certification processes, statistical process control, Six Sigma methodology. Every auto component plant in Pune has dozens of them. Odisha’s mineral processing plants have a handful.
Process control engineers — specialists who monitor and optimize continuous production lines using SCADA systems, PLCs, and process analytics. A modern steel rolling mill is essentially a software-controlled mechanical system. The engineers who run it are as much programmers as they are metallurgists.
Automation and PLC programmers — the people who write the control logic for industrial robots, conveyor systems, and process equipment. India’s automation workforce is concentrated in Pune, Chennai, and Bangalore. Odisha produces near-zero.
Die and tool designers — the specialists who create the molds, dies, jigs, and fixtures that shape metal into finished products. This is a craft that takes years to master and exists only where stamping and forming industries exist. It does not exist in Odisha.
Supply chain managers — people who coordinate the movement of materials, components, and finished products across complex manufacturing networks. Mining supply chains are simple (ore to truck to rail to port). Manufacturing supply chains are webs of hundreds of suppliers, each delivering specific components on precise schedules. Managing them is a discipline. Odisha’s economy does not generate demand for it.

The National Skill Development Corporation (NSDC) skill gap studies estimate that Odisha needs 2-3 million skilled industrial workers by 2030 to meet its stated development targets. The state’s current training infrastructure produces a small fraction of that number. And of what it produces, a large share leaves.

This is not a gap that can be closed with a policy announcement. It is a structural deficit that has accumulated over decades, compounding in the way that all capability deficits compound: the absence of industry means the absence of skills, which means the absence of the conditions that attract industry, which deepens the absence of skills. Breaking this cycle requires intervention at a scale that Odisha has not yet attempted (confidence level: high — approximately 85% — based on publicly available state skill development budgets and ITI expansion plans through FY 2025-26, which show incremental rather than transformative investment).

The Ganjam Paradox

Everything discussed so far is abstract compared to what happens in Ganjam district, which provides the single most powerful illustration of Odisha’s labor economics — and the single most powerful refutation of the claim that Odisha lacks an industrial workforce.

An estimated 7 lakh (700,000) people from Ganjam district work in Surat, Gujarat. This is not a rough guess. It is a widely cited figure corroborated by multiple sources — IndiaSPEND, Scroll.in, the Aajeevika Bureau, and Gujarat government estimates. The corridor has existed for over 40 years as a large-scale industrial migration route, and over 80 years if you count its earlier, smaller incarnation. It is one of the largest single-district-to-single-city migration corridors in India.

What do these 700,000 people do in Surat? They operate power looms. An estimated 98 percent of the workforce in Surat’s textile power loom industry is from Odisha. They run 1.5 million loom machines across Surat, producing the synthetic textiles — polyester, nylon, blended fabrics — that clothe much of India and supply export markets from Africa to Southeast Asia. Surat’s textile industry is worth tens of thousands of crores annually. Its workforce is almost entirely Odia.

These workers are not doing unskilled labor. Power loom operation requires training, dexterity, and the ability to manage complex machinery running at high speeds. The workers maintain the looms, troubleshoot breakdowns, manage yarn feeds, and maintain quality standards. Beyond textiles, Odias in Surat work in the diamond polishing industry — cutting and polishing rough diamonds with a precision measured in fractions of a millimeter. Over 90 percent of the world’s polished diamonds are produced in Surat. The skill required is extraordinary. The training takes months.

These people acquired industrial skills. They acquired them in Surat, not in Odisha. They migrated as unskilled agricultural laborers from Ganjam’s villages. They learned to operate power looms, manage yarn inventory, assess fabric quality, and troubleshoot machinery on the job, in Gujarat. The skills they possess — machine operation, quality control, production discipline, factory-floor teamwork — are precisely the skills that any manufacturing economy needs.

The paradox is this: Odisha says it cannot industrialize because it lacks skilled labor. Meanwhile, 700,000 skilled industrial workers from one Odisha district are running the textile industry of another state.

The labor is not missing. The factories are.

If Surat’s textile factories were in Berhampur — Ganjam’s main city, 600 kilometres from Surat — those 700,000 workers would not have migrated. They would be doing the same work, acquiring the same skills, producing the same output. The difference would be where the economic value accrues: to Ganjam’s local economy instead of Surat’s.

The remittances partially quantify what is lost. Ganjam-to-Surat remittances are estimated at Rs 120 crore per month — roughly Rs 1,440 crore per year, and this is likely an undercount since much money travels through informal channels. Some estimates of total annual remittances from Odia workers in Surat to their home districts put the figure at Rs 3,000-5,000 crore per year. This money sustains villages, pays for children’s education, funds medical treatment, and builds houses. It is economically significant. But it is a fraction of the value those workers create. The factory owner’s profit, the commercial rent, the local taxes, the supplier ecosystem revenue, the induced service economy — all of that stays in Surat. Ganjam gets the remittance check. Gujarat gets the industrial economy.

The working conditions deserve mention, not for emotional effect but because they reveal the economic terms on which this labor is supplied. Workers operate 12-hour shifts earning Rs 20,000-25,000 per month. Sound levels inside loom units exceed 110 decibels. A 2018 Aajeevika Bureau medical examination found 95 percent of workers reported hearing loss. Workers live in rooms of 500-800 square feet, packed 60-100 workers per room across two shifts. Between 2012 and 2015, 84 fatal events killed 114 workers and seriously injured 375 in registered textile processing units. Eight workers died from electrocution while operating looms in the May-August 2024 period alone. In 25 years of work in certain areas, as Scroll.in documented, not a single unit provides all workers with identity cards, records overtime, or reports accidents.

These conditions exist because the workers have no alternative. The alternative is Ganjam, where the factory does not exist. The rational economic calculation — 12-hour shifts in deafening noise at Rs 25,000 per month in Surat, versus Rs 10,000-15,000 per month as an agricultural laborer or mining daily-wager in Odisha — is not a real choice. It is a structural compulsion dressed as individual agency.

In investing terms, consider what is happening. An asset (the labor force) is being deployed at below its potential return because the portfolio (the state economy) is concentrated in a single low-return asset class (mining and agriculture). The workers themselves are diversifying by migrating — they are, individually, rational investors seeking higher returns. But they are doing it at enormous personal cost (separation from family, health damage, exploitative conditions), and the returns accrue to a different portfolio (Gujarat’s economy). If the home portfolio were diversified — if manufacturing existed in Odisha — the same workers could earn the same or higher returns without the migration cost. The system-level inefficiency is staggering.

The Full Migration Map

Ganjam-to-Surat is the most dramatic corridor, but it is not the only one. Odisha’s labor exports are extensive.

Kerala construction labor: Hundreds of thousands of Odias work in construction across Kerala — Kochi, Thiruvananthapuram, Kozhikode, and smaller towns. Kerala’s own population is aging, its fertility rate has dropped below replacement level (1.6 children per woman), and Keralites increasingly refuse manual labor. The gap is filled by workers from Odisha, Bihar, and West Bengal. Odia construction workers in Kerala earn Rs 18,000-30,000 per month — significantly more than they would earn in equivalent work in Odisha, where construction activity is lower and pay rates reflect the weaker local economy.

Hyderabad services: A significant Odia population works in cleaning, security, domestic services, and increasingly in IT services in Hyderabad. The IT professionals tend to be NIT/IIT graduates from Odisha who settled in Hyderabad’s tech corridor. The service workers are from poorer districts.

Bangalore IT: An estimated 600,000 Odias live in Bangalore, driven by the IT boom of the late 2000s and the general migration of educated south-eastern Indians toward India’s tech capital. NIT Rourkela, VSSUT Burla, IIT Bhubaneswar, and Odisha’s engineering colleges supply a steady stream of graduates to Bangalore’s IT companies, consultancies, and startups.

The dadan system — brick kilns and bonded labor: From western Odisha — Bolangir, Nuapada, Kalahandi, Bargarh — an estimated 60,000 or more families migrate annually to brick kilns in Andhra Pradesh, Telangana, Tamil Nadu, and Chhattisgarh. Unlike the Surat corridor, which is voluntary (if coerced by economic necessity), the dadan system involves debt bondage. Workers take wage advances from contractors called sardars or khatadars, typically during Nuakhai festival or to fund medical emergencies, and then are obligated to work at distant brick kilns until the debt is repaid. Sixty percent of inter-state migration from western Odisha goes to brick kilns. It is family migration — men, women, and children. The ILO has documented it. It is, by any reasonable definition, a form of modern unfree labor.

Gulf countries: From coastal Odisha — Balasore, Bhadrak, Puri, Ganjam — workers migrate to the construction booms of Saudi Arabia, UAE, Qatar, and other Gulf states. The numbers are harder to quantify because they are embedded in the larger Indian labor migration to the Gulf, but the pattern is well-established.

Total Odia out-migration: The Periodic Labour Force Survey (PLFS) 2020-21 estimated that approximately 8.51 lakh (851,000) laborers from Odisha migrate annually for work. This is certainly an undercount. Adding longer-term migrants who have settled semi-permanently in other states — the IT professionals in Bangalore, the textile workers who have been in Surat for decades, the construction workers in Kerala who return only for festivals — the total is likely in the range of 20-30 lakh people (2-3 million) working outside the state at any given time. Some estimates, using E-Shram portal registrations and other data, put the total number of Odia migrant workers even higher.

For a state of 46 million people, 2-3 million working-age adults absent from the local economy is not a demographic footnote. It is a structural feature. It means that 33 percent of males aged 18-35 are absent from the local workforce, according to one estimate. It means villages in Ganjam where, for much of the year, only women, elderly, and children remain. It means an economy that has effectively exported its working-age population to generate value for other states.

The Wage Staircase — Why People Leave

Put the numbers together and a staircase emerges — a clear, rational hierarchy of economic returns that explains, without any need for cultural or psychological theories, why Odisha’s workers leave.

Role	Location	Monthly income (Rs)	Working conditions
Agricultural laborer	Ganjam / Bolangir	6,000-10,000	Seasonal, rain-dependent, minimal
Mining daily laborer	Keonjhar / Angul	10,000-15,000	Dangerous, dusty, inconsistent
Brick kiln worker	AP / Telangana	12,000-18,000	Bonded, family labor, exploitative
Textile worker	Surat	20,000-25,000	Indoor, steady, deafening, long hours
Construction worker	Kerala	18,000-30,000	Outdoor, physically demanding, steady
Steel plant technician	Jamshedpur	30,000-50,000	Industrial, skilled, benefits
Auto component machinist	Pune / Chennai	25,000-40,000	Factory, skilled, growth path
IT professional (entry level)	Bangalore	40,000-80,000	Office, comfortable, growth path

Every step up in skill and value addition is a step out of Odisha.

Read the table from top to bottom. The lowest-paying work is located in Odisha. The highest-paying work is located elsewhere. A rational worker — any rational economic actor — looks at this table and leaves. The worker is not irrational. The worker is not culturally predisposed to migration. The worker is not displaying some deficiency of loyalty. The worker is responding to price signals that are entirely clear: the return on labor is higher outside Odisha than inside it, at every skill level above subsistence agriculture.

The problem is not the worker. The problem is the wage structure that a mining-and-agriculture economy creates.

In portfolio theory, this is a concentration risk problem. Odisha’s economy is like a portfolio concentrated in a single low-return asset class — primary extraction. The “returns” (wages) are low because extraction is the least value-creating stage of the commodity chain. The workers are, individually, diversifying by moving their labor to higher-return asset classes (manufacturing in Gujarat, services in Bangalore, construction in Kerala). But they are doing this at high transaction costs (migration, family separation, exploitative conditions). If the home portfolio were diversified — if manufacturing and services existed at scale within Odisha — the weighted average return (average wage) for the entire workforce would increase, and the transaction costs of migration would be eliminated.

This is not a controversial claim. It is arithmetic. The controversy, such as it is, lies in why the diversification has not happened. That question involves capital, policy, infrastructure, and institutional capacity — subjects of other chapters. Here, the point is narrower: the labor gap is not a cause of Odisha’s underdevelopment. It is a symptom. The workers exist. The skills can be built (and have been built, in Surat, in Kerala, in Jamshedpur). What does not exist is the industrial structure that would employ them at home.

What Would It Take to Close the Gap

Consider a scenario — not a prediction, but a thought experiment grounded in concrete numbers.

India’s National Steel Policy targets 300 MTPA of crude steel capacity by 2030-31 (a target that will almost certainly be missed, but the direction is clear). Odisha’s current crude steel capacity is approximately 22 MTPA, against iron ore production of 130-140 MTPA. Suppose Odisha doubled its steel capacity to 44 MTPA over the next decade — absorbing a significantly larger share of its own ore.

The workforce implications of that doubling:

Direct industrial employment: An additional 50,000-80,000 jobs in steel plants and associated processing facilities. These are not mining jobs at Rs 10,000-15,000 per month. They are industrial jobs — technicians, operators, engineers, quality inspectors, maintenance specialists — at Rs 25,000-80,000 per month. The weighted average wage in the additional jobs would be approximately 2-4 times the average mining wage.

Indirect and induced employment: Applying the 3-4x multiplier documented for steel plant investment, an additional 150,000-250,000 jobs in supplier industries, service providers, logistics, education, healthcare, and commerce. These indirect jobs create the middle class that a mining economy cannot produce — the shopkeepers, teachers, doctors, mechanics, restaurateurs, and entrepreneurs who form the economic tissue of a functioning industrial city.

Total employment impact: 200,000-330,000 additional jobs, most at wages significantly above current mining labor rates.

What would it take to build the workforce for this scenario?

Skill infrastructure:

10-20 new ITIs focused specifically on metallurgy, machining, welding, automation, and quality control. Not general-purpose ITIs producing generic fitters and electricians, but industry-specific training centers with curricula designed in partnership with steel companies and auto component manufacturers.
Significant expansion of NIT Rourkela’s metallurgical and materials science capacity — more faculty, more lab space, more research partnerships with industry.
Industry-linked training programs modeled on the German dual-system apprenticeship model, where trainees split time between classroom instruction and factory-floor experience. Germany’s apprenticeship system is the reason its manufacturing workforce remains among the world’s best despite high labor costs. The model has been adapted in other countries — South Korea, Singapore, parts of China. It could be adapted for Odisha, but it requires industry partners willing to invest in training, which requires industries that actually exist.

Timeline: Building a critical mass of skilled industrial workers — not just the first generation, but the experienced supervisors, the seasoned engineers, the shop-floor veterans who transmit tacit knowledge — takes 10-15 years under the best conditions. This is not a guess. It is what the development of every successful industrial cluster in India has taken. Pune’s auto cluster did not spring into existence overnight. It evolved from the 1960s and 1970s, when Bajaj and Tata moved operations there, through decades of workforce development, supplier ecosystem growth, and institutional accumulation. Jamshedpur’s steel ecosystem took a century.

Fifteen years is a long time for a state whose working-age population is leaving now. This creates a chicken-and-egg problem that is genuinely difficult to solve. Industry will not invest without a skilled workforce. The workforce will not develop without industry to employ them. Breaking the deadlock requires someone — government, anchor investors, or some combination — to absorb the initial mismatch cost.

The AI wildcard — a tentative assessment (confidence level: moderate, approximately 60%):

There is a possibility that AI and automation could compress the timeline. If AI-assisted manufacturing reduces the skill barrier for certain operations — if a process control system with machine learning can compensate for an operator’s lack of experience, or if computer vision can perform quality inspections that currently require years of trained human judgment — then the workforce development timeline might shrink from 15 years to 8-10 years. This is speculative, and anyone claiming certainty about AI’s impact on manufacturing workforce development is selling something. But the possibility exists and is worth flagging.

What AI does not do is eliminate the need for workers. It changes what they do. Fewer blast furnace operators, more process control technicians. Fewer manual quality inspectors, more data analysts monitoring sensor streams. Fewer traditional machinists, more CNC programmers. The skill profile shifts upward — which is why the ITI and engineering college capacity question remains central even in an AI-augmented manufacturing scenario. The workers still need training. The training content changes.

This would be wrong if AI progress in manufacturing slows significantly (due to the physical world being harder to automate than the digital world, a real possibility), or if India’s manufacturing sector fails to adopt AI-assisted processes at scale (due to cost, institutional resistance, or inadequate digital infrastructure). It would also be wrong if the skill-reducing effects of AI are overstated — if, in practice, AI-assisted manufacturing still requires experienced human operators who can intervene when the system fails in novel ways. The history of automation suggests that the last 20 percent of human skill is the hardest to replace, and that attempts to fully automate complex physical processes frequently underperform.

The Structural Inversion

Step back and see the full picture.

Odisha produces the raw materials. Other states process them. Odisha’s educational institutions produce the engineers. Other states employ them. Odisha’s districts produce the workers. Other states absorb them. In every dimension — material, human capital, labor — the flow is outward.

This is not a natural state of affairs. It is the result of specific historical decisions (freight equalization, central allocation of industrial licenses, nationalization of coal, underinvestment in rail and port infrastructure), specific policy failures (the gap between announced investments and materialized factories), and the compounding effects of those decisions over seven decades. But the result, regardless of its causes, is a structural inversion: the state that has the most reason to industrialize — because it has the raw materials, the population, and the fiscal incentive — is the state where industrialization has least occurred.

The labor gap is not a problem to be solved in isolation. It is a symptom of a system that extracts value from Odisha at every level — mineral value, human capital value, and labor value. Closing the gap requires not just more ITIs or more engineering seats, but a fundamental shift in the state’s position within the national value chain. More processing within the state means more jobs within the state. More jobs means more skills development, because skills develop in response to demand, not in anticipation of it. More skills means more processing becomes viable. The cycle can turn virtuous. But first, it has to be turned.

The 700,000 Odias in Surat are not evidence that Odisha lacks industrial capacity. They are evidence that Odisha’s industrial capacity is deployed in the wrong state. If the conditions were created for even a fraction of that demonstrated capability to be exercised at home, the labor gap would begin to close — not because new workers would be conjured from nowhere, but because the workers who already exist would have a reason to stay.

Sources

National Institute of Technology Rourkela. “Department of Metallurgical and Materials Engineering — Placement Statistics.” NIT Rourkela, 2023-24. https://www.nitrkl.ac.in/MME
Mahanadi Coalfields Limited. “Annual Report 2023-24.” MCL, Sambalpur. https://www.mahanadicoal.in
NALCO (National Aluminium Company Limited). “Annual Report 2023-24.” NALCO, Bhubaneswar. https://nalcoindia.com
Tata Steel Limited. “Integrated Report and Annual Accounts 2023-24 — Jamshedpur Operations.” Tata Steel, Mumbai. https://www.tatasteel.com
JSPL (Jindal Steel and Power Limited). “Annual Report 2023-24 — Angul Operations.” JSPL, New Delhi. https://www.jindalsteelpower.com
SAIL (Steel Authority of India Limited). “Rourkela Steel Plant — Performance Overview 2023-24.” SAIL, New Delhi. https://sail.co.in
Tata Steel Limited. “Kalinganagar Expansion Project — Progress Update 2024.” Tata Steel, Mumbai. https://www.tatasteel.com
IndiaSPEND. “How Caste Identity Prevails Among Odia Migrant Workers In Surat.” IndiaSPEND, 2024. https://www.indiaspend.com/welfare/how-caste-identity-prevails-among-odia-migrant-workers-in-surat-936127
Scroll.in. “In Surat’s grimy living quarters for Odia migrants, caste is the dividing line.” Scroll.in, 2024. https://scroll.in/article/1077340/in-surats-grimy-living-quarters-for-odia-migrants-caste-is-the-dividing-line
Aajeevika Bureau. “Workers in Surat’s Textile Power Loom Industry — Health and Living Conditions Survey 2018.” Human Dignity Foundation. https://www.humandignity.foundation/wp-content/uploads/2018/11/Odisha-State-Migration-Profile-Report.pdf
People’s Archive of Rural India (PARI). “Synthetic fabric, authentic despair.” PARI. https://ruralindiaonline.org/en/articles/synthetic-fabric-authentic-despair/
Ministry of Labour and Employment, Government of India. “Periodic Labour Force Survey (PLFS) 2020-21 — Annual Report.” https://mospi.gov.in
National Skill Development Corporation (NSDC). “District Skill Gap Study — Odisha.” NSDC, New Delhi. https://nsdcindia.org
Directorate General of Training, Ministry of Skill Development and Entrepreneurship. “Industrial Training Institutes — State-wise Statistics.” Government of India. https://dgt.gov.in
Indian Institute of Technology Bhubaneswar. “School of Minerals, Metallurgical and Materials Engineering — Overview.” IIT Bhubaneswar. https://www.iitbbs.ac.in
Scroll.in. “How Ganjam became Odisha’s Covid-19 hotspot.” Scroll.in, 2020. https://scroll.in/article/969139/how-ganjam-and-not-capital-bhubaneshwar-became-odishas-covid-19-hotspot
Gram Vikas / UNDP / CMID. “Labour Migration from Rural Odisha — Jagannathprasad Block, Ganjam.” 2020-21. https://www.gramvikas.org/wp-content/uploads/2021/09/Jagannathprasad-Block-Migration-Profile-Final-Web-9-Sept-21.pdf
Citizen Matters. “Living in rooms by looms: Fate of migrant workers in Surat.” https://citizenmatters.in/surat-odisha-looms-migrant-workers/
The Wire. “Despite Large Numbers, Surat’s Migrant Power Loom Workers Remain Politically Ignored.” https://m.thewire.in/article/politics/despite-large-numbers-surats-migrant-power-loom-workers-remain-politically-ignored
IDR (India Development Review). “How migration is changing villages in Odisha.” https://idronline.org/article/livelihoods/how-migration-is-changing-villages-in-odisha/
Odisha Plus. “Migrant multiplier: How Odisha’s youth are powering India’s informal economy.” 2025. https://odisha.plus/2025/06/migrant-multiplier-how-odishas-youth-are-powering-indias-informal-economy/
OdishaBytes. “Migration: Odias Are Makers of New India; Why It’s Not a Matter of Pride.” https://odishabytes.com/migration-odias-are-makers-of-new-india-why-its-not-a-matter-of-pride/
ILO (International Labour Organization). “India Employment Report 2024.” ILO, New Delhi. https://www.ilo.org
Ministry of Steel, Government of India. “National Steel Policy 2017 — Updated Targets.” https://steel.gov.in
Work Fair and Free. “Study of Migration from Ganjam District, Odisha to Gujarat.” Working Paper 01. https://workfairandfree.org/wp-content/uploads/2024/08/Working-Paper-01_Final.pdf
The Migration Story. “Of caste, climate and the creation of one of India’s busiest migration corridors.” https://www.themigrationstory.com/post/of-caste-climate-and-the-creation-of-one-of-india-s-busiest-migration-corridors
Organiser. “Odisha Dadan Migration: Patterns, Response and Strategic Intervention of the New BJP Govt.” 2025. https://organiser.org/2025/08/26/311088/bharat/odisha-dadan-migration-patterns-response-and-strategic-intervention-of-the-new-bjp-govt/
The Federal. “Migrant distress shadows Nuakhai.” 2024. https://thefederal.com/category/features/western-odisha-nuakhai-labour-trafficking-kbk-region-kalahandi-202403
NewsClick. “Surat: Cost of a Billion-Dollar Textile Industry.” https://www.newsclick.in/surat-cost-billion-dollar-textile-industry

Source Research

The raw research that informs this series.