English only · Odia translation in progress

Chapter 7: What a City Needs

In 2019, a software engineer named Prateek — NIT Rourkela, Class of 2015, four years at a mid-sized product company in Bangalore — decided he wanted to come home. Not on a visit. Home to stay. He had a plan: a hyperlocal food delivery startup for Bhubaneswar. The logic was sound. Bhubaneswar’s population was crossing a million. The city had a growing IT workforce. Swiggy and Zomato were present but thin — three or four delivery partners for an entire neighborhood, forty-minute wait times, half the restaurants not listed. The gap was real. He had savings, a co-founder, and the technical chops to build the platform.

Within six months, Prateek was back in Bangalore.

The reasons, when he listed them on a Reddit thread that circulated among Odia tech workers, read like a diagnostic report on a failed operating system. First: addresses. Bhubaneswar has no standardized addressing system for most residential areas. Delivery partners could not find houses. Google Maps pins were frequently 200 to 500 meters off because the underlying map data was wrong. He spent weeks manually correcting coordinates for the first 300 restaurant partners. Second: roads. The main arteries were fine. The last-mile roads — the ones delivery partners actually ride on — were potholed, unlit, and in many neighborhoods, unpaved during monsoon. Average delivery time was 55 minutes against his target of 30. Third: payment infrastructure. UPI worked, but the restaurant owners overwhelmingly preferred cash. His unit economics required 70% digital payments to avoid cash-handling costs. He was getting 35%. Fourth: cold chain. No warehouse-grade refrigerated storage was available for rent at any price point he could afford. The one cold storage facility in the city was designed for agricultural produce, not restaurant-grade inventory. Fifth: regulatory clarity. He spent three months trying to understand the licensing requirements for a food delivery aggregator in Odisha. The FSSAI process was clear at the national level but the municipal requirements were opaque. Nobody at BMC could tell him definitively what permits he needed.

Each of these failures — addresses, roads, payments, logistics, regulation — is, in software terms, a missing system service. Not the application. The operating system underneath it. Prateek did not fail because his idea was bad or his team was weak or the market did not exist. He failed because Bhubaneswar does not yet have the urban operating system on which a business like his can run.

This chapter is about that operating system — what it consists of, why it matters, what Odisha’s cities are missing, what other states built, and the single counter-example that suggests Odisha knows how to build one when it chooses to.

[Note: Prateek’s story is a composite drawn from multiple accounts by Odia entrepreneurs on Reddit and Quora who attempted startups in Bhubaneswar between 2018 and 2024. The specific details — addressing failures, delivery time gaps, cold chain absence — are documented across these accounts. The composite is used to protect individual identities while preserving the pattern. Confidence in the pattern: high, ~85%. Confidence in any single attributed detail: moderate, ~65%.]

The Platform Concept

Edward Glaeser, the Harvard economist who has spent three decades studying why cities exist, boils it down to a single mechanism: proximity creates productivity. When people cluster together, three things happen that do not happen when they are dispersed. First, knowledge spills over — the apprentice steelworker learns from the master, the junior developer picks up coding patterns from the senior, the street food vendor watches the competitor’s technique and adapts it overnight. Second, labor markets pool — an employer can find specialized workers, and a specialized worker can find employers, because both are concentrated in the same place. Third, inputs are shared — the logistics company, the legal firm, the accounting practice, the restaurant that feeds the office workers, the daycare that watches their children — all become economically viable because density provides enough customers.

Glaeser calls these agglomeration effects. They are the reason cities exist despite being expensive, noisy, polluted, and stressful. The productivity gains from proximity outweigh the costs. His research demonstrates that a worker in a city that is twice as dense as another earns, on average, 6% more — not because of different skills but because of the environment. Doubling the density of an area raises the productivity of workers by approximately 6 to 28 percent depending on the industry. The effect is strongest in knowledge-intensive industries, which is why IT companies cluster in Bangalore rather than distributing themselves evenly across Karnataka.

Jane Jacobs, writing sixty years before Glaeser had the econometrics to prove it, identified the conditions under which these agglomeration effects actually emerge. Her four conditions for city vitality, published in The Death and Life of Great American Cities in 1961, remain the most durable framework in urban planning:

One: mixed uses. A district must serve more than one primary function — residential, commercial, industrial, institutional — to ensure people are present at different hours for different purposes. A purely residential neighborhood empties during working hours. A purely commercial district dies at night. Mixed use keeps the streets alive, which keeps them safe, which keeps them attractive, which sustains the density that generates agglomeration.

Two: short blocks. The street fabric must be fine-grained — frequent corners, multiple routes between any two points. Long blocks create dead zones. Short blocks create possibilities for encounter. This sounds like urban design trivia until you realize what it means economically: every corner is a potential storefront, every intersection is a potential market.

Three: buildings of varying age and condition. This is the one that confuses people. Jacobs was not arguing for historical preservation. She was arguing for economic diversity. New buildings have high rents that only established, profitable businesses can afford. Old buildings have low rents where new businesses can start, immigrants can open shops, artists can find studios, non-profits can operate. A district of all-new buildings is a district of chain stores and corporate offices. A district with a mix of old and new is a district with economic variety — and variety is what generates the unexpected combinations that create new value.

Four: sufficient density. Not maximum density. Sufficient density. Enough people per acre to generate the foot traffic, the customer base, the labor pool, and the random encounters that Jacobs called “the intricate sidewalk ballet.” Too little density and nothing reaches critical mass. Too much density without infrastructure creates the slum. The range matters.

Now here is the conceptual move that matters for Odisha. Both Glaeser and Jacobs are describing the same underlying reality: a city is not a collection of buildings. A city is a platform — an environment that enables activities that could not occur without it. The buildings are hardware. The roads, pipes, wires, and rails are peripheral devices. But without the operating system — the layer of organized services that sits between the hardware and the human activity — nothing runs.

A computer without an operating system is a box of silicon. You can have the best processor in the world, the fastest memory, the most advanced graphics card. Without an OS that manages memory allocation, schedules processes, handles input and output, manages file storage, and provides a networking stack, no application can execute. The hardware is there. The capability is there. But the capability is latent, unrealized, useless — because the layer that translates capability into function is missing.

A city’s operating system has six layers. Each is necessary. None is sufficient alone. And critically: they must work together, not as independent projects. A city with excellent transport but dysfunctional land markets is like a computer with a perfect networking stack but a corrupted file system. Technically, some things work. Practically, the platform is broken.

The six layers:

Layer 1: Land markets. Accessible, transparent, functional. The ability to buy, sell, lease, and develop land through predictable processes with clear titles. This is the file system — where everything is stored and accessed.

Layer 2: Transport. Connecting people to jobs to markets to each other. This is the networking layer — how data moves between nodes.

Layer 3: Talent retention. The pipeline from education to employment to retention. Not just producing skilled people but keeping them. This is the CPU — processing capacity.

Layer 4: Governance. Predictable rules, efficient services, accountable institutions. This is the kernel — the core of the operating system that every other service depends on.

Layer 5: Infrastructure services. Water, power, waste management, drainage, digital connectivity. These are the background system services — you notice them only when they fail.

Layer 6: Network effects. Each new person or business makes the platform more valuable for every existing person and business. This is the self-reinforcing dynamic that turns a collection of buildings into a city. It is what makes the whole exceed the sum of the parts.

Every Indian city that functions as a genuine economic platform — Bangalore, Hyderabad, Pune, Chennai, Ahmedabad, Surat — has all six layers operational, even if imperfectly. Every Odia city that fails to generate self-sustaining economic activity is missing at least three. The rest of this chapter examines each layer in Odisha and explains why the operating system was never installed.

Layer 1: Land Markets — The Missing Foundation

Every operating system needs a file system. Without it, you cannot store data, retrieve it, organize it, or share it. The file system is so fundamental that most users never think about it — until it breaks, and then nothing works. Land markets are the file system of a city. Without functional land markets, you cannot allocate space to its highest-value use, which means you cannot build density, which means you cannot generate agglomeration, which means the city never becomes a platform.

Odisha’s land market dysfunction begins underground, in the records themselves. The Bhulekh portal — the state’s digital land records system, built under the National Land Records Modernization Programme that began in 2008 — represents genuine progress. Citizens can look up Records of Rights online, view plot-level ownership details, and check mutation status. This is real. But the World Bank’s Land Governance Assessment Framework report on Odisha documented what every Odia who has attempted to buy or sell land already knows: the records are riddled with inaccuracies. Overlapping claims. Un-updated mutations. Mismatches between revenue records and survey maps. In peri-urban areas — the zones where cities actually grow — the gap between what the records say and what exists on the ground is a chasm.

The conversion process makes it worse. Under the Odisha Land Reforms (Amendment) Act, 1960, agricultural land can be converted to non-agricultural use only with prior permission from the Tahasildar or Sub-Collector. The process involves application, local inspection, verification, fee payment, and issuance of a conversion certificate. On paper, this is orderly. On the ground, it is widely understood to be slow, opaque, and vulnerable to rent-seeking. Peri-urban land conversion — the mechanism through which every Indian city actually expands — routinely happens informally before formal conversion, creating a grey zone of semi-legal construction that is technically unauthorized but practically irreversible.

RERA Odisha (ORERA), established in October 2017, has registered over 19,000 projects and 1,200 agents. But the organized real estate market — the market that ORERA actually regulates — covers a fraction of total construction activity. Bhubaneswar’s developer ecosystem is dominated by local, mid-sized firms: Utkal Builders (established 1990), Archid Builders (established 2008), Metro Group, Motwani Constructions. No national-scale developer — DLF, Godrej Properties, Prestige — has significant presence. The reason is not lack of demand. It is the absence of clear title chains, the difficulty of land assembly, and the regulatory opacity that makes large-scale organized development too risky for firms with other options.

Now compare this with what Gujarat built. The Town Planning Scheme mechanism — used in Ahmedabad since the early twentieth century and scaled across Gujarat since the 1970s — is a fundamentally different approach to urban land management. In a TPS, agricultural land on the urban fringe is pooled by the development authority. The authority plans the pooled area — laying out roads, parks, utility corridors, and plots. It returns approximately 50 to 60 percent of the original land to the farmers as planned, serviced urban plots. The remaining 40 to 50 percent goes to the development authority for roads, parks, utilities, and sale plots (which fund the infrastructure). The farmer who owned an acre of agricultural land worth Rs 20 lakh ends up owning a smaller but fully serviced urban plot worth Rs 80 lakh. The city gets planned expansion with infrastructure built in from the start. The authority funds itself through the process.

Gujarat has executed over 400 Town Planning Schemes. Ahmedabad’s entire modern expansion — the western suburbs that house millions of people — was built through TPS. The mechanism solved three problems simultaneously: land assembly without adversarial acquisition (farmers participate because they profit), infrastructure provision without separate budgetary allocation (funded through the value uplift), and market creation (clear titles on planned plots enable a transparent real estate market).

Odisha has executed essentially zero Town Planning Schemes. The Development Authorities — BDA, CDA, RDA, SDA — exist as planning bodies, but their primary mechanism for land development is government land allocation, not market-based land pooling. BDA controls 556 revenue villages covering Bhubaneswar, Khurda, and Jatni. Its Comprehensive Development Plan for 2040 envisions accommodating a projected population of 30 lakh, requiring 4.31 lakh new dwelling units and Rs 31,450 crore in investment. These are planning numbers. The mechanism for translating them into actual serviced land in the hands of actual buyers through a transparent market process does not exist.

Without a functional land market, the first OS layer is missing. Everything built on top is built on an unstable foundation. When an IT company evaluates Bhubaneswar and finds that assembling a 50-acre campus requires negotiating with thirty different landowners, each with a title chain of uncertain validity, through a conversion process of uncertain duration, subject to clearances of uncertain requirements — that company chooses Hyderabad, where HMDA (Hyderabad Metropolitan Development Authority) can offer a ready, clear-titled plot in an approved IT park zone. The company’s decision has nothing to do with Odisha’s talent pool or climate or culture. It has to do with the file system. If the file system is corrupted, you cannot save your work.

Layer 2: Transport — The Single-City Solution

Mo Bus launched in November 2018 in Bhubaneswar. Since then, it has become one of Indian urban transit’s genuine success stories. The numbers: 560-plus buses including 180 electric vehicles. Twenty-two routes covering the Bhubaneswar urban area. Average daily ridership of approximately 91,706 passengers. A 200 percent increase in ridership over its first four-and-a-half years. Fifty-seven percent of commuters shifted from private modes — meaning more than half the people riding Mo Bus previously used personal vehicles, auto-rickshaws, or two-wheelers. The service won recognition as India’s first city-level transit system launched entirely by a state government, as opposed to central-government-funded metro projects.

This is real. Mo Bus works. The Capital Region Urban Transport body (CRUT) that operates it has demonstrated that when Odisha builds genuine urban infrastructure, citizens respond. The demand was always there. The supply was missing.

Now expand the frame. Cuttack, twenty-five kilometers from Bhubaneswar, population approximately 700,000, has historically had no organized public transport. (Partial service has recently begun expanding from Bhubaneswar.) Rourkela, population 311,000, steel city, home to NIT Rourkela — no public transport system. Berhampur, population 356,000, largest city in southern Odisha — no public transport system. Sambalpur, population 204,000, gateway to western Odisha — no public transport system. Balasore, Baripada, Puri outside of partial Mo Bus extension, Jharsuguda, Paradip — none of them have organized bus services.

The expansion plan exists on paper. CRUT has announced expansion to Cuttack, Puri, Rourkela, Sambalpur, and Berhampur. Some of this is beginning to materialize. But the current reality, as of early 2026, is that organized public transit in Odisha is functionally a Bhubaneswar phenomenon.

Compare this with Tamil Nadu. The Tamil Nadu State Transport Corporation (TNSTC) operates over 22,000 buses connecting every town in the state. Not just Chennai. Salem, population 917,000, has state-run bus services. Erode, population 521,000, has state-run bus services. Tirunelveli, population 474,000, has state-run bus services. Towns with populations of 50,000 have TNSTC coverage. The Chennai Metro Rail is expanding. Suburban rail carries 1.2 million passengers daily. The Mass Rapid Transit System runs along the coast. The transport layer in Tamil Nadu is not a single-city solution. It is a statewide platform.

Karnataka: BMTC (Bangalore Metropolitan Transport Corporation) operates approximately 6,400 buses in Bangalore. KSRTC provides statewide services. Namma Metro is expanding with Phase 2 under construction. Suburban rail is being built. Even Mysore, the second city, has its own KSRTC city service.

Maharashtra: BEST in Mumbai (3,400 buses), PMPML in Pune (1,900 buses), city bus services in Nagpur, Nashik, Aurangabad. Mumbai’s suburban rail network — the lifeline — carries 7.5 million passengers daily. Mumbai Metro is expanding. Pune Metro is operational.

Gujarat: Ahmedabad’s Janmarg Bus Rapid Transit System operates a 160-kilometer network carrying 349,000 daily riders, widely cited as a model for Indian cities. Surat, Vadodara, and Rajkot all have city bus services. Ahmedabad Metro is operational.

The pattern is unmistakable. Every state that urbanized successfully built transport as a statewide platform, not a capital-city project. The transport layer connects people to jobs, jobs to markets, markets to suppliers, suppliers to inputs. Without it, the urban operating system has no networking stack. Data cannot flow between nodes. Each city operates in isolation, unable to benefit from the connectivity that makes urban systems productive.

Bhubaneswar’s Metro Rail project — Phase 1, 26 kilometers from Trisulia near Cuttack to Biju Patnaik International Airport, estimated cost Rs 6,256 crore, target completion December 2027 — will add capacity to the one city that already has the most transport infrastructure. This follows the pattern. Investment goes where infrastructure already exists, reinforcing the concentration rather than building the network. Approximately Rs 460 crore had been spent by late 2024, with soil testing at 10 percent completion and depot construction beginning at Ratagada and Malipada. Delhi Metro Rail Corporation is floating tenders for nineteen elevated stations.

The metro will help Bhubaneswar. But Odisha’s transport problem is not Bhubaneswar’s congestion. Odisha’s transport problem is that a worker in Sambalpur cannot commute to a job 15 kilometers away without a personal vehicle. A student in Berhampur cannot reach a coaching center in the next town without hiring an auto-rickshaw. A small manufacturer in Rourkela cannot move goods to Jharsuguda without private trucking. The networking layer is absent across the entire state except for one node.

Inter-city connectivity compounds the problem. Bhubaneswar’s Biju Patnaik International Airport is the only functional commercial airport in the state. It now has direct international flights to Bangkok, Dubai, and Singapore (launched in May 2023), and a skybridge connecting Terminal 1 and 2 was inaugurated in February 2024. Jharsuguda has limited domestic service. Rourkela has seasonal or weekly flights. No airport serves southern Odisha — the Ganjam-Koraput corridor from which 700,000 people migrate annually to Surat alone. No high-speed rail connects any Odia city to any other Odia city. The Talcher-Bimlagarh railway line, planned for over seventy years, remains incomplete.

In operating system terms: the networking stack exists in one node. Every other node is running on dial-up — or offline entirely.

Layer 3: Talent — The Export Machine

Odisha produces engineers, doctors, managers, and scientists at rates that would be the envy of most developing countries. The institutional infrastructure is formidable: IIT Bhubaneswar, NIT Rourkela, KIIT University (30,000 students across programs), IIM Sambalpur, AIIMS Bhubaneswar, NISER (National Institute of Science Education and Research), Xavier University (XIMB). The combined annual output runs into the thousands of skilled professionals.

The retention rate for top performers is near zero.

NIT Rourkela’s placement data for 2024-25 tells the story in a single table. The institute reported 1,274 placement offers from companies including Amazon, Microsoft, Goldman Sachs, Texas Instruments, and Schlumberger. Virtually all placements were outside Odisha — Bangalore, Hyderabad, Pune, Delhi, Mumbai, Chennai. The highest offered CTC was approximately Rs 63 lakh per annum (for an international role). The average package for the top tier was Rs 20-25 lakh. No employer in Odisha — not one — was offering packages in this range during the placement season.

KIIT University’s 2024 placement season was even more telling because of scale. The university reported 5,585 offers from companies including Amazon, McKinsey, Salesforce, Deloitte, JP Morgan, and Goldman Sachs. These are global firms recruiting at a Bhubaneswar-based university. They are recruiting because KIIT produces the talent they need. They are placing the recruited talent in Bangalore, Hyderabad, Pune, and overseas — because that is where the runtime environment exists.

This is the OS metaphor made precise. Odisha has an excellent compiler. The education system takes raw talent — eighteen-year-olds from Ganjam, Cuttack, Sambalpur, Koraput — and compiles it into skilled professionals who are immediately employable at the highest levels of the global economy. The compiler works. It has been refined over decades. It produces output of internationally competitive quality.

But a compiler is not a runtime environment. The compiler converts source code into executable form. The runtime environment is where the executable actually runs — the memory management, the process scheduling, the I/O handling, the libraries and frameworks that the executable calls upon. Without a runtime environment, the compiled code cannot execute. It sits on a disk, perfect but useless.

Odisha’s compiled code — its engineering graduates, its MBA holders, its medical doctors, its research scientists — cannot execute in Odisha because the runtime environment does not exist. What constitutes the runtime? Jobs at competitive compensation. A critical mass of peer talent that creates the intellectual stimulation knowledge workers need. Career progression paths that do not require relocating. Quality-of-life infrastructure — housing, healthcare, education for children, entertainment, cultural life — that makes staying attractive. And network effects: the presence of other talented people and firms that creates opportunities no individual job can provide.

Bangalore has all of this. That is why 6 lakh Odias live there, by some estimates. Hyderabad has it. Pune has it. Chennai has it. Bhubaneswar has the IT companies — TCS, Infosys, Wipro, Accenture, Cognizant, Tech Mahindra, IBM, Genpact, Capgemini, EY, Deloitte are all present. But presence is not ecosystem. Most of these operations in Bhubaneswar are back-office or delivery centers, not product development hubs. The work is lower-value. The compensation is lower. The career ceiling is lower. A software engineer who starts at TCS Bhubaneswar and wants to work on cutting-edge AI will be transferred to Bangalore or Hyderabad within two years — or will leave.

The economic arithmetic is brutal. Odisha’s per capita GSDP is approximately Rs 1.47 lakh (FY2023-24 estimate). Karnataka’s is approximately Rs 3.27 lakh. Maharashtra’s is approximately Rs 3.08 lakh. Tamil Nadu’s is approximately Rs 3.16 lakh. A knowledge worker in Bhubaneswar is operating in an economy that generates roughly half the per-capita output of the economies where their peers work. That gap translates directly into compensation, career opportunity, and quality of life. No amount of nostalgia for home can overcome a two-to-one income differential sustained over a thirty-year career.

The result is what economists call a brain drain equilibrium. Odisha invests in producing talent. The talent leaves. The economy that produced it does not benefit from the agglomeration effects that talent would generate if it stayed. Because the talent leaves, the economy does not develop the services and opportunities that would retain talent. Because the economy does not develop, more talent leaves. The equilibrium is stable and self-reinforcing. Breaking it requires changing the economics — not by asking people to sacrifice income out of state loyalty, but by building the urban operating system that makes staying economically rational.

This is not hypothetical. It is the story of Bangalore itself. In the 1980s, Bangalore was a pensioner’s paradise — pleasant climate, slow pace, military and aerospace establishments. The IT industry did not choose Bangalore because Bangalore was already a tech hub. It chose Bangalore because the Karnataka government, through its 1997 IT Parks Policy, created the institutional framework — dedicated infrastructure zones, tax concessions, single-window clearances — that made Bangalore the lowest-friction location for IT companies. Texas Instruments opened its Bangalore center in 1985. Infosys was founded there in 1981 (originally Pune, relocated to Bangalore in 1983). Once a critical mass formed, the agglomeration effects took over. Each new company made Bangalore more attractive for the next one. The runtime environment became self-sustaining.

Odisha’s challenge is not to replicate Bangalore. It is to build a runtime environment where the talent its excellent compiler produces can execute — even if the initial applications are modest. The alternative is to continue subsidizing the training of workers for other states’ economies. It is an expensive form of generosity.

Layer 4: Governance — The Capacity Gap

The kernel of an operating system is the innermost layer — the code that manages memory, schedules processes, handles hardware interrupts, and provides the fundamental services that everything else depends on. If the kernel crashes, the entire system goes down. If the kernel is slow, every application is slow. If the kernel has bugs, they propagate unpredictably into every service that depends on it. You can have the best applications in the world. If the kernel is broken, they will not run.

Municipal governance is the kernel of a city’s operating system. Property tax collection funds the services. Planning departments manage land use. Engineering departments build and maintain infrastructure. Health departments ensure sanitation. The municipal government is the entity that maintains the platform.

Odisha’s municipal kernel is technically installed but barely functional.

The numbers are stark. The state has approximately 176 Urban Local Bodies as of early 2026 — 7 Municipal Corporations, roughly 59 Municipalities, and approximately 110 Notified Area Councils. The expansion from approximately 114 ULBs to 176 through 24 new NACs and 5 new municipalities announced in December 2025 acknowledged the reality of peri-urban growth. Whether administrative capacity followed the notification is the question.

Property tax — called “holding tax” in Odisha — is the primary own-revenue source for ULBs. The rates are 17.5 percent of annual value for municipalities and 20.5 percent of annual rental value for Municipal Corporations. These rates sound reasonable until you learn the collection reality. The CAG’s 2023 performance audit, covering 2015-16 to 2019-20, found that only 29 to 42 percent of allocated funds were actually spent by municipalities. Not collected — spent. Odisha’s urban local bodies could not use even the money they had.

That number deserves a pause. Not 29 to 42 percent revenue collection — that would be bad but explicable as a tax enforcement problem. Twenty-nine to forty-two percent expenditure of allocated funds. The money was available. The municipalities could not absorb it. They lacked the staff to plan projects, the technical capacity to execute them, the administrative systems to process payments, and in many cases the basic institutional competence to spend allocated budgets. This is not a revenue problem. It is a capacity problem.

The consequences were direct. During the Fourteenth Finance Commission period (FY 2017-20), Odisha’s ULBs forfeited Rs 286.25 crore out of Rs 354.51 crore in performance grants — grants specifically conditioned on meeting urban reform benchmarks. The municipalities failed to meet the conditions. Not because the conditions were impossible. Because the municipalities lacked the capacity to implement the reforms that would have triggered the grants. Free money was left on the table because the kernel could not process the request.

Staffing deficits explain much of the capacity gap. CAG audit reports identify vacancy rates of approximately 25 percent for environment engineers, 20 percent for health inspectors, 20 percent for sanitation staff, and severe shortages of town planners that the reports flag but cannot quantify because the positions were often never sanctioned in the first place. In smaller municipalities and NACs, the deficit is absolute: many lack even a single qualified engineer or town planner.

The 74th Constitutional Amendment, enacted in 1992 and effective from June 1993, was supposed to fix this. The Amendment mandated genuine urban self-governance by requiring regular ULB elections, reservations for SC/ST and women, devolution of 18 functions listed in the Twelfth Schedule (including urban planning, regulation of land use, roads and bridges, water supply, public health, slum improvement, and urban poverty alleviation), Ward Committees, District Planning Committees, and State Finance Commissions.

Odisha’s formal compliance is exemplary. The state is among only 9 states that formally devolved all 18 Twelfth Schedule functions to ULBs. On the Urban Governance Index, Odisha ranked first nationally with a score of 56.86, including first place in the “Empowered Citizens” category.

Ranked first. On paper.

The CAG’s performance audit revealed what “formal devolution” means in practice:

Of the 18 devolved functions: 1 function has no ULB role at all. 3 functions have ULBs acting merely as implementing agencies for state-designed programs. 1 function has a confused dual role between ULB and state. And 13 functions — the vast majority — have minimal ULB role or overlapping state jurisdiction that effectively neuters municipal authority.

Thirteen of eighteen. The functions were devolved on paper. The staff, budgets, technical expertise, and political autonomy required to actually perform those functions were not transferred. What Odisha built is the governance equivalent of installing an operating system that reports all services as running while actually running none of them. The system monitor says everything is green. The user experience is that nothing works.

This is the hollow institution pattern documented across the SeeUtkal research: formal structures that look complete from outside but lack the internal capacity to function. It appears in PESA implementation (the law was adopted; 136 CAG violations documented in practice). It appears in panchayat devolution (powers transferred on paper, funds and functionaries withheld). It appears in land records (digitized but inaccurate). And it appears in municipal governance: first on the governance index, last on governance delivery.

The kernel is installed. But it runs in a degraded state, handling interrupts late, dropping processes, corrupting memory. Every application that depends on it — every business that needs a building permit, every resident who needs a water connection, every developer who needs a plan approval, every entrepreneur who needs regulatory clarity — experiences the kernel’s dysfunction as their own failure. They do not think “the municipal governance system is broken.” They think “doing business in Bhubaneswar is difficult.” They are describing the same thing.

Layer 5: Infrastructure Services — The Two-Hour Supply

System services run in the background. On a well-functioning computer, you do not think about memory management, disk defragmentation, network routing, or process scheduling. They happen. You open an application, it opens. You save a file, it saves. You send a message, it sends. The system services are invisible precisely because they work.

In Odisha’s cities, system services are visible because they fail. Routinely.

Start with water. Bhubaneswar’s official supply figure is 248 LPCD — liters per capita per day. This sounds adequate. The national benchmark is 135 LPCD. Bhubaneswar supplies nearly double the standard. Case closed, service delivered, move on.

Except the water arrives for approximately two hours per day. Morning and evening, intermittent supply. The 248 LPCD is a production figure — what the treatment plants produce and push into the network. The distribution figure — what actually reaches households and for how long — is a fundamentally different story. Massive distribution losses (pipe leakage, unauthorized connections, system inefficiencies) mean that what is produced at the plant and what flows from the tap at the end of the pipe are different numbers. Twenty-four-by-seven water supply — the standard in every developed city and the stated goal of Odisha’s urban water policy since 2013 — remains a goal, not a reality.

The “Drink from Tap” project — an initiative to provide potable tap water — has faced criticism for inconsistent delivery. The State Urban Water Supply Policy of 2013 mandates at least 70 LPCD for all urban residents. For cities smaller than Bhubaneswar, even this modest standard is not reliably met.

In operating system terms: this is a system that crashes and reboots on a schedule. You know it will be available between 6 and 7 AM and between 5 and 6 PM. You plan your life around the crashes. You buy water tanks for the intervals. You install pumps to pressurize what trickles through the pipes. You adapt to the dysfunction instead of the system adapting to your needs. A two-hour water supply is technically functional in the same way that a computer that needs to be manually restarted every hour is technically functional. You can use it. You cannot build anything serious on it.

Sanitation and waste management are worse. A CAG audit found that not a single sanitary landfill exists in the entire state of Odisha. Every gram of municipal solid waste generated by every city in the state goes to open dumping grounds. Bhubaneswar Municipal Corporation is working with GIZ (German development agency) on construction and demolition waste management and improved collection systems, but the baseline is zero scientific disposal infrastructure across the state.

Stormwater drainage coverage in Bhubaneswar — ranked Smart City Number One in India’s initial selection round in January 2016 — is approximately 45 percent. More than half the city has no formal stormwater drainage. In flood-prone areas, 53 to 65 meters of drains per stretch are prone to choking from solid waste, because the waste management system and the drainage system fail together. When it rains, the Smart City Number One goes underwater. The monsoon flooding is an annual event documented by local media with the inevitability of a recurring software bug that nobody fixes because the underlying architecture is wrong.

Sewage treatment capacity is a national problem, not just an Odia one — India as a whole treats approximately 28 percent of its sewage. But in a state where municipal capacity is already strained to the point of leaving performance grants unclaimed, the sewage gap is unlikely to close faster than the national average.

The bright spot is electricity, and it is worth understanding why. Odisha’s power distribution, privatized through a joint venture with Tata Power in 2020 across four distribution companies (TPCODL, TPSODL, TPNODL, TPWODL), has seen aggregate technical and commercial losses drop to 22.6 percent in the first half of FY2025, with double-digit reduction since the partnership began. TPCODL earned an A+ rating and ninth rank nationally among distribution utilities. Smart meters: 8.34 lakh installed. Substations automated: 753. The company became the first Indian power utility to integrate DigiPIN, mapping over 93,000 transformers and 19 lakh electricity poles with digital address codes.

Why did power improve while water, waste, and drainage did not? Because power distribution was handed to an entity — Tata Power — with the institutional capacity, technical expertise, capital, and operational systems to execute. The private entity brought its own “kernel” — its own management systems, its own staffing, its own operational discipline. It did not depend on municipal governance capacity. It operated outside the broken kernel. This is not an argument for blanket privatization. It is an observation about institutional capacity: when a system service is operated by an entity with functional management, it improves. When it is operated by a municipal system with 20-25 percent staff vacancies and 29-42 percent fund utilization, it does not.

Digital infrastructure is the one area where India’s national-level platform — Jio’s 4G/5G rollout, Airtel’s network expansion, the BharatNet fiber program — has leapfrogged state-level dysfunction. Odisha has 3.32 crore mobile subscribers, 2.21 crore mobile broadband subscribers, 5G coverage across all 30 districts (Airtel) and 13 cities (Jio), with teledensity at 80 percent against a national average of approximately 85 percent. This infrastructure was built by private companies responding to national-level market incentives and spectrum auctions. The state government was not the builder. But the infrastructure is there, and it works, and it is one of the few layers of the urban operating system in Odisha where the gap with developed states is narrow and closing.

The infrastructure services layer, in sum: water supply intermittent, waste disposal non-existent in sanitary terms, drainage covering less than half the capital city, sewage treatment inadequate, power improving through privatization, digital infrastructure improving through national-level market forces. The system services that depend on municipal governance are failing. The system services that operate outside municipal governance are working. The diagnosis writes itself.

What Other States Built

A city’s operating system does not build itself. In every Indian state where cities function as economic platforms, the state government made a deliberate, sustained, institutional investment in urban infrastructure that went beyond the capital city. Not Smart City projects — those are application-layer investments on top of an OS that is assumed to already exist. Not individual schemes or missions — those address specific symptoms. What the successful states built was the platform itself, through dedicated institutions that operated at state scale, over decades.

Tamil Nadu: The Ecosystem Builder

Tamil Nadu’s urban story is not Chennai. Chennai matters, obviously — it is India’s fourth-largest city, with a metropolitan population exceeding 11 million, the automobile capital (“Detroit of India”), an IT hub second only to Bangalore in the south. But what makes Tamil Nadu different from Odisha is not Chennai. It is Coimbatore and Madurai and Tiruchirappalli and Salem and Tirunelveli.

Coimbatore grew from a textile town to a diversified manufacturing hub — textiles, pump manufacturing, auto components, IT services — with a metropolitan population of 2.15 million. It did not grow because the state government built it. It grew because the platform existed for it to grow on. SIPCOT (State Industries Promotion Corporation of Tamil Nadu, established 1971) created 24 industrial complexes across 15 districts, 6 Special Economic Zones, covering 35,043 acres. SIPCOT’s IT Park in Siruseri is Asia’s largest. Critically: SIPCOT operates across the state, not just in Chennai. It built the industrial infrastructure that allowed organic clusters to form in secondary cities.

TNSTC — the state transport corporation — runs 22,000-plus buses connecting every town. Chennai has its metro, suburban rail, and MRTS (Mass Rapid Transit System). But the transport platform is statewide. A worker in Erode can commute to a factory in Tiruppur. A student in Salem can reach a university in Coimbatore. The connectivity layer does not stop at the capital’s municipal boundary.

Municipal governance in Tamil Nadu benefits from institutional depth. The Chennai Corporation, established in 1688, is one of India’s oldest municipal bodies. The Directorate of Town and Country Planning provides technical support. The CMDA (Chennai Metropolitan Development Authority, established 1975) has a track record of master plan preparation that extends decades. These are not perfect institutions. But they exist, they are staffed, and they function at a level that allows the governance kernel to process basic requests.

Tamil Nadu’s model: state-created industrial infrastructure (SIPCOT) distributed across the state, plus organic cluster development supported by that infrastructure (Coimbatore textiles, Hosur manufacturing, Tiruppur knitwear), plus comprehensive statewide transport (TNSTC and Chennai transit), plus institutional municipal governance built over centuries.

Gujarat: The Infrastructure-First Approach

Gujarat’s urbanization story is inseparable from two institutions: GIDC and the Town Planning Scheme.

GIDC (Gujarat Industrial Development Corporation) created a dense network of industrial estates across the state. Not in Ahmedabad alone. Across the state — Surat, Vadodara, Rajkot, Bhavnagar, Morbi, Anand, Vapi, Ankleshwar. The estates provided ready-to-use industrial plots with power, water, roads, and drainage. They lowered the barrier to entry for manufacturing. A small entrepreneur who wanted to start a ceramics unit in Morbi or a diamond-polishing workshop in Surat did not need to navigate land acquisition, utility connections, and regulatory clearances from scratch. The platform existed. The entrepreneur needed only to build the application.

The Town Planning Scheme, as discussed in the land markets section, solved the urban expansion problem. Gujarat has executed 400-plus TPS, enabling planned city growth without the adversarial land acquisition that paralyzes urbanization in other states. The mechanism is so well-established that Ahmedabad Municipal Corporation became India’s first ISO-certified municipal body — not because ISO certification is inherently meaningful, but because the institutional capacity required to achieve it reflects a governance kernel that actually functions.

The Vibrant Gujarat summit, started in 2003, operates as an investment coordination mechanism — a regular, institutionalized process for connecting investors with opportunities. Whether the announced investment figures materialize fully is debatable (they often do not). But the summit serves a platform function: it creates a predictable, recurring event around which investment decisions coalesce. It is a scheduled process in the operating system — running every two years, handling a defined set of inputs and outputs.

Gujarat’s model: dense industrial infrastructure (GIDC) distributed statewide, plus innovative land management (TPS) enabling planned urban growth, plus public transport (Ahmedabad BRTS, metro), plus strong organic clusters that grew on the platform (Surat diamonds and textiles, Morbi ceramics, Rajkot engineering, Ankleshwar chemicals).

Karnataka: The Sector-Specific Platform

Karnataka’s approach was narrower and more deliberate than Tamil Nadu’s or Gujarat’s. The state government, recognizing in the 1990s that Bangalore had the raw ingredients for an IT industry (educated workforce, English proficiency, pleasant climate, defense and aerospace establishments), passed the 1997 IT Parks Policy. This policy created the institutional framework — dedicated IT infrastructure zones with guaranteed power, water, and telecom; tax concessions; single-window clearances — that lowered the friction for IT companies to set up operations.

The result: ITPB (International Tech Park Bangalore) in Whitefield, Electronics City, Manyata Tech Park, Embassy Golf Links — dedicated technology campuses that provided the platform on which the IT ecosystem grew. KIADB (Karnataka Industrial Areas Development Board) operated across the state, not just in Bangalore. The state attempted a deliberate Mysore-as-second-city strategy, directing some IT investment to the heritage city.

Namma Metro, now expanding significantly in Bangalore, and suburban rail under construction are the transport investments catching up to the employment growth that the IT platform generated. The sequence was: build the sector-specific platform first, let the employment and population follow, then build the transport infrastructure. It is not the ideal sequence (infrastructure should precede growth), but it worked because the initial platform was compelling enough to overcome the infrastructure deficits.

Karnataka’s model: deliberate sector-specific policy (IT Parks) creating the initial platform, plus KIADB industrial areas across the state, plus metro and transit investment following the growth, plus the leveraging of existing city qualities (Bangalore’s climate, talent pool, and military-academic base).

Maharashtra: The Multi-City Machine

Maharashtra’s urban platform is the most diversified. MIDC (Maharashtra Industrial Development Corporation) has built 289-plus industrial areas across the state. Not in Mumbai and Pune alone. Across Maharashtra — Nashik, Aurangabad, Nagpur, Kolhapur, Solapur. MMRDA (Mumbai Metropolitan Region Development Authority) handles metropolitan coordination for the Mumbai region. But the state-level platform operates beyond Mumbai.

Pune’s dual-sector model — IT (Hinjewadi IT Park) and automobiles (Pimpri-Chinchwad manufacturing corridor) — demonstrates what happens when the platform supports multiple industries. Pune grew from 2.5 million in 1991 to over 7 million by 2021, driven by the interaction between the IT and manufacturing ecosystems. Each sector generated demand for the other’s workers, suppliers, and services. The agglomeration effects multiplied.

Maharashtra’s model: organic financial hub (Mumbai) built over centuries, plus deliberate industrial diversification (Pune IT + auto, Nashik wine + manufacturing, Aurangabad auto), plus massive statewide industrial infrastructure (MIDC), plus strong municipal governance tradition (BMC is one of India’s richest municipal corporations).

The Common Thread

What Tamil Nadu, Gujarat, Karnataka, and Maharashtra share is not a single policy or a single institution. What they share is a state-level commitment to building the urban platform across multiple cities, sustained over decades, through dedicated institutions with the mandate and capacity to execute.

SIPCOT operates across 15 Tamil Nadu districts. GIDC covers all of Gujarat. KIADB serves all of Karnataka. MIDC has 289 areas across Maharashtra.

Odisha has IDCO — the Odisha Industrial Infrastructure Development Corporation, established in 1981. IDCO’s mandate includes industrial infrastructure. It has built multi-product MSME parks at seven locations: Ramdaspur (Cuttack), Chhatabar (Khurda), Talamulasasan (Angul), Basantpur (Sambalpur), Reuna (Sundargarh), Kesinga (Kalahandi), and Barapali/Hinjili (Ganjam). It plans new MSME parks and IDCO towers. But IDCO functions primarily as a land provider, not an ecosystem builder. Providing a plot is necessary. It is not sufficient. The difference between IDCO and SIPCOT is the difference between offering a parking space and building a highway. Both involve land. One creates connectivity; the other does not.

The urban platform question is not “does Odisha have institutions?” It does. The question is whether those institutions operate at the scale, intensity, and sustained commitment that platform-building requires. The evidence suggests they do not — with one extraordinary exception.

The Operating System Odisha Doesn’t Have

Pull the layers together. Assess each one honestly.

Land markets: Bhulekh digitization is genuine progress. But land records remain inaccurate in peri-urban areas. Conversion processes are opaque. No Town Planning Scheme mechanism exists. RERA covers a fraction of construction activity. The organized real estate market is thin. National developers are absent. Assessment: the file system is partially digitized but fundamentally unreliable. Score: barely functional.

Transport: Mo Bus is a genuine success in Bhubaneswar. Expansion to other cities is beginning but far from complete. No city outside Bhubaneswar (and partial Cuttack/Puri extensions) has organized public transit. Metro is for Bhubaneswar only. One functional commercial airport in the entire state. Assessment: the networking layer exists in one node. Score: single-city solution.

Talent retention: World-class educational institutions producing thousands of graduates annually. Near-zero retention of top performers. IT companies present in Bhubaneswar but primarily as delivery centers, not product hubs. Two-to-one income gap with major metro cities. Assessment: excellent compiler, no runtime environment. Score: net exporter.

Governance: 176 ULBs. First rank on Urban Governance Index. All 18 functions formally devolved. But 13 of 18 functions with minimal ULB role in practice. 29-42% fund utilization. Rs 286 crore in performance grants forfeited. 20-25% staff vacancies. Assessment: kernel installed but running in degraded mode. Score: formally devolved, practically incapacitated.

Infrastructure services: Two-hour water supply in the capital. No sanitary landfill in the state. 45% stormwater drainage in Bhubaneswar. Power improving through Tata Power partnership. Digital infrastructure improving through national-level private investment. Assessment: system services fail routinely except where operated outside municipal governance. Score: intermittent.

Network effects: Without the other five layers functioning, network effects cannot emerge. Each missing layer reduces the value the platform offers, which reduces the incentive for new people and businesses to join, which prevents the positive feedback loop that turns a collection of buildings into a city. Assessment: absent. Score: not yet initiated.

This is not a failing grade on individual tests. It is the absence of the operating system itself. No applications can run. IT companies cannot scale because they cannot find enough talent willing to stay (Layer 3) in a city where the governance kernel is broken (Layer 4) and infrastructure services are intermittent (Layer 5). Entrepreneurs cannot start because the land market is opaque (Layer 1), transport is limited (Layer 2), and regulatory clarity does not exist (Layer 4). Professionals cannot stay because the quality of life — the user interface of the platform — does not compete with Bangalore or Pune or Hyderabad.

The operating system was never installed.

This would be a counsel of despair except for one thing: Odisha has demonstrated, in one specific domain, that it knows exactly how to build a platform when it chooses to. And the design principles it used are precisely the ones the rest of the urban operating system needs.

JAGA: The Counter-Narrative and the Design Principle

JAGA Mission — the Odisha Liveable Habitat Mission — began on August 8, 2017, when the State Cabinet approved the Odisha Land Rights to Slum Dwellers Act. The numbers: 2,919 slums across 114 cities. 175,000 families granted land tenure security. Approximately 1.7 million slum dwellers reached. 707 slums fully upgraded to liveable habitats. 2,724 slums with 100 percent piped water connections. 8 cities declared slum-free. Two World Habitat Awards — Bronze in 2019 and Bronze again in 2023. A case study at the University of Chicago Law School. Recognition from UN-Habitat and Metropolis.

These numbers matter, but the design principle matters more.

Every other major Indian slum rehabilitation program treats slums as a housing problem. Mumbai’s Slum Rehabilitation Authority (SRA) model: give free flats to slum dwellers in exchange for granting additional FSI (Floor Space Index) to developers. Telangana’s GHMC model: mix of rehabilitation housing and relocation. Karnataka’s BDA/BBMP model: mix of relocation and in-situ development. The common assumption: the problem is that people lack adequate shelter. The solution: provide shelter.

JAGA inverts this. JAGA treats slums as a land rights problem first and a housing problem second. The sequence matters enormously:

Step 1: Grant land titles. End the eviction risk. Give slum dwellers legal ownership of the land they occupy. This single act — transferring a piece of paper — transforms the family’s relationship with the state, with financial institutions, and with their own investment calculus. A family with title will invest in improving their house because they know they will not be evicted. A family without title will not, because any investment can be bulldozed.

Step 2: Upgrade infrastructure in situ. Do not relocate communities to distant housing projects where they lose their social networks, their proximity to work, and their community structures. Bring the infrastructure — piped water, toilets, electricity, storm drainage, paved streets, lighting, open spaces, children’s play areas, community centers — to where the people already are.

Step 3: Enable self-construction through PMAY subsidies and micro-credit. Families build homes suited to their needs, in their existing communities, on land they legally own. The state provides the subsidy and the framework. The family provides the agency.

Step 4: Hand maintenance to the community. Upgraded facilities are managed by the community, creating ownership and sustainability.

This is correct operating system design. In software engineering, you build the kernel before the applications. You establish the file system before you write files. You set up memory management before you allocate memory. JAGA got the sequence right: it secured the base layer first (land rights — the file system) before building features (housing, infrastructure — the applications). The base layer makes everything above it possible and sustainable. Without it, the features are unstable — built on land that could be reclaimed, in communities that could be evicted, with investments that could be destroyed.

The design is philosophically closer to Hernando de Soto’s thesis in The Mystery of Capital than to conventional Indian urban policy. De Soto argued that the poor in developing countries possess enormous assets — but in defective forms. Their houses are built on land with no clear title. Their businesses operate outside the law. Their assets cannot be used as collateral, cannot be traded in formal markets, cannot be used to generate capital. Formalizing these assets — making the implicit explicit, the informal formal — is what unlocks economic potential. JAGA is de Soto’s thesis implemented at scale, for 175,000 families across 114 cities.

And here is the critical observation: JAGA was not a Bhubaneswar-only solution. It operated across 114 cities. This is the one major urban initiative in Odisha that broke the capital-city pattern. It reached Sambalpur and Berhampur and Rourkela and Balasore and every municipality and NAC in the state. It demonstrated that the state government has the institutional capacity to design, fund, and execute a statewide urban platform initiative when political will and institutional design align.

If JAGA’s design logic — secure the base layer first, then build features, and operate at state scale not city scale — were applied to the entire urban operating system, it would produce something like this:

First, fix the land market across the state. Not just slum titles — though those matter. Complete land records digitization with accurate cadastral mapping. Implement transparent conversion processes. Consider Town Planning Schemes for peri-urban areas in every city above 200,000 population. Create the file system.

Second, build the transport platform statewide. Extend Mo Bus/Ama Bus to every city above 200,000. Create inter-city connectivity. Build the networking layer.

Third, build the talent retention infrastructure. Transform IT delivery centers into product development hubs. Create internship-to-employment pipelines connecting KIIT, NIT Rourkela, IIT Bhubaneswar, and IIM Sambalpur to local industry. Develop sectors beyond IT — food processing, mineral value-addition, maritime services — that create diverse employment. Build the runtime environment.

Fourth, build municipal governance capacity. Professional staffing at scale. Genuine financial autonomy. Skill-building for elected representatives. Fix the kernel.

Fifth, make infrastructure services reliable. Twenty-four-hour water supply. Sanitary waste disposal. Complete drainage. Build on what Tata Power demonstrated: that competent institutional management produces results. Stabilize the system services.

This is what Gujarat and Tamil Nadu did over decades. Not in a single policy announcement. Not in a five-year plan. Over decades of sustained institutional investment, led by state-level bodies with the mandate and capacity to operate across multiple cities, funded through creative mechanisms (TPS in Gujarat, SIPCOT’s self-financing model in Tamil Nadu), and sustained across political transitions.

The question for Odisha is not capability. JAGA proved the capability exists. OSDMA proved it exists. Mo Bus proved it exists. The state can build world-class systems when it focuses. The question is whether it can build an urban operating system — all six layers, across all its cities, sustained over decades — rather than building individual applications on a platform that does not exist.

This is a question of institutional design, not political will alone. Political will is a moment. Institutional design is a structure. Moments pass. Structures endure. What JAGA built is a structure — a legal framework, an implementation mechanism, a scaling model. What the urban operating system needs is the same thing at ten times the scope.

I would estimate, with approximately 55 to 60 percent confidence, that Odisha’s current institutional trajectory — concentrating urban investment in Bhubaneswar through Smart City and Metro projects while under-investing in statewide platforms — will produce a city-state pattern rather than an urbanized state. Bhubaneswar will become a 5-million-person metropolis by 2040 or 2050, with functional infrastructure, a growing IT sector, and reasonable quality of life. It will coexist with a state where Sambalpur, Berhampur, Rourkela, and every other city remain essentially what they are today: medium towns without the platform to grow. The talent export will continue. The 17 percent will rise to perhaps 22 or 25 percent — driven almost entirely by Bhubaneswar’s growth and the reclassification of peri-urban areas. The urban system will remain structurally shallow.

This would be wrong if: the state government creates a dedicated urban platform institution with statewide mandate (an “Odisha Urban Development Corporation” on the SIPCOT/GIDC model), adequately funded and staffed, operating across all major cities, with legal authority to implement Town Planning Schemes and build industrial infrastructure. It would be wrong if Mo Bus expansion reaches every city above 200,000 by 2030. It would be wrong if NIT Rourkela’s placement data in 2030 shows even 15 percent of graduates taking jobs in Odisha. Any one of these signals would indicate that the operating system is being installed rather than just the applications.

The compound interest of this matters. Every year the operating system is not installed, the gap with Tamil Nadu, Gujarat, Karnataka, and Maharashtra widens. Their urban platforms generate network effects that attract more talent, more capital, more businesses, which strengthens the platform, which attracts more. Odisha’s talent continues to compile and export. The compiled code runs on someone else’s operating system and generates value for someone else’s economy. The loss is not just the talent that leaves. It is the agglomeration effects that would have been generated if that talent had stayed — the knowledge spillovers, the labor market pooling, the input sharing that Glaeser describes. These effects are non-linear. They compound. And they compound on the platform where the talent executes, not where it was compiled.

A city is an operating system that provides services to applications. Without the OS layer, no applications can run, regardless of how talented the developers or how good the hardware. Odisha has the developers. It has considerable hardware. It does not have the operating system. JAGA proved it knows how to build one. The question is whether it will.

Sources

Agglomeration Economics and Urban Theory:

  • Glaeser, Edward. “Agglomeration Economics.” NBER Working Paper, Chapter 1. National Bureau of Economic Research. Available at: https://www.nber.org/system/files/chapters/c7977/c7977.pdf
  • Jacobs, Jane. The Death and Life of Great American Cities. Random House, 1961.
  • De Soto, Hernando. The Mystery of Capital: Why Capitalism Triumphs in the West and Fails Everywhere Else. Basic Books, 2000.

Municipal Governance Data:

Transport Data:

Land Markets and Real Estate:

Water, Sanitation, and Infrastructure:

Power and Digital Infrastructure:

Comparator State Institutions:

JAGA Mission:

Economic Data:

Cross-References

The Churning Fire (Consciousness Series):

  • Chapter 3: “The One Who Sees First” — institutional design and the platform-builder archetype. The consciousness-shifter as someone who sees the system-level problem, not just the symptoms.
  • Chapter 8: “What Remains” — institutional architecture and compound interest of platform investments. OSDMA as proof of dormant capacity, relevant to the question of whether Odisha can build sustained urban institutions.

Political Landscape Series:

  • Governance capacity analysis. The 74th Amendment implementation gap documented here extends the hollow institution pattern identified across municipal governance. Municipal weakness as a specific expression of the broader governance-on-paper-not-in-practice dynamic.

The Long Arc (Ninety-Year Transformation):

  • Chapter 3: “The Cathedral in the Village” — Rourkela Steel Plant as cathedral dropped in tribal district without ecosystem, directly parallel to Bhubaneswar as urban investment dropped without statewide platform.
  • Chapter 5: “The Extraction Equilibrium” — the Naveen model (extract, welfare, votes, extract) and its implications for whether institutional capacity-building (which has no immediate electoral payoff) can compete with welfare disbursement (which does).
  • Chapter 7: “What Persists, What Was Solved, What Broke” — urbanization identified as a “failed” category in the ninety-year scorecard. This chapter explains why.

The Leaving (Migration and Diaspora):

  • Chapter 1: “The Numbers and the Names” — the scale of talent export (2 to 5 million Odias outside the state). This chapter explains the structural reason: no runtime environment for compiled talent.
  • Chapter 4: “The Skilled Departure” — NIT Rourkela placement data, IT sector brain drain to Bangalore. The talent layer analysis here deepens that chapter’s observations.
  • Chapter 8: “The Return Ticket” — structural conditions for return migration. This chapter identifies the specific platform layers that would need to exist for return to become economically rational.

Women’s Odisha (The Invisible Half):

  • Chapter 6: “The Network That Already Exists” — Mission Shakti as platform infrastructure. JAGA Mission and Mission Shakti share a design principle: build the base layer (land rights / financial inclusion) before building features (housing / economic activity).
  • SHG as institutional platform — parallel to the argument that Odisha can build platforms when institutional design and political will align.

Environmental Odisha (The Foundation Shifts):

  • Chapter 5: “The Coast That Holds” — infrastructure-environment intersection. Urban drainage failures during monsoon are an environmental infrastructure problem.
  • Chapter 1: “The Storm That Changed Everything” — OSDMA as the model for what focused institutional capacity can achieve. The urban platform question is whether OSDMA’s design logic can be applied to slower, less dramatic challenges.

Value Chain Series:

  • Chapter 3: “How Others Built It” — SIPCOT, GIDC, KIADB comparisons. The industrial ecosystem analysis in this chapter parallels and extends that chapter’s treatment of state-level development institutions.
  • Chapter 5: “The Labor Gap” — the workforce economics that connect to the talent retention layer. No mineral value-addition industry can develop without a skilled workforce that stays.

Tribal Odisha (The Parallel Civilisation):

  • Chapter 5: “The Mountain and the Mine” — JAGA’s land rights approach contrasted with the land alienation experienced by tribal communities. Both are about the foundational question of who owns the land, but the trajectories diverge: urban slum dwellers gained rights while tribal communities continued to lose them.

Source Research

The raw research that informs this series.