English only · Odia translation in progress

Chapter 7: The Cyclone and the Cheque

On the night of October 29, 1999, the Bay of Bengal sent Odisha a message that would reshape the state’s relationship with Delhi more thoroughly than any election, any policy announcement, or any Finance Commission formula ever had. A super cyclone --- classified as the equivalent of a Category 5 hurricane, with sustained wind speeds exceeding 260 kilometres per hour and a storm surge that pushed the ocean seven metres inland --- struck the coast near Paradip, Jagatsinghpur district, and proceeded to destroy everything in its path for the next thirty-six hours.

The numbers that emerged after the winds stopped are the kind that render normal governance vocabulary inadequate. At least 9,885 people dead, according to the state government’s white paper. Other estimates, including the EM-DAT disaster database, placed the toll at 10,915. Some researchers suggested the real number could have been as high as 30,000, with unregistered deaths in isolated villages never counted. More than 70 percent of the dead had drowned --- not in the ocean, but in the storm surge that turned the coastal plain into a temporary sea. Approximately 13 million people were affected across fourteen districts. Around 1.67 million houses were damaged or destroyed --- 23,129 washed away entirely, 746,337 fully destroyed, 880,620 partially damaged. Over 18,420 square kilometres of cropland was submerged. An estimated 444,000 livestock perished. The total economic damage: US$4.44 billion.

Into this devastation, the Government of India sent a cheque. Prime Minister Atal Bihari Vajpayee announced the allocation of Rs 3 billion --- approximately US$69.3 million --- to the Odisha state government. This supplemented an earlier allocation of US$59.5 million made for the less severe cyclone that had struck the same coast two weeks earlier.

The Odisha state government had asked for Rs 5 billion for immediate relief and Rs 25 billion for rehabilitation. It received one-tenth of its demand. The total central allocation of roughly Rs 3 billion against total damage of approximately Rs 20,000 crore tells you everything about how Delhi values Odisha when the cameras are rolling and the floodwaters are rising.

This is not a chapter about a natural disaster. This is a chapter about what a disaster reveals --- about who responds, how much they give, who gets the relief, who doesn’t, and what happens after the cameras leave.

The Arithmetic of Abandonment

Start with the gap, because the gap is the story.

Odisha demanded Rs 30 billion total --- Rs 5 billion for relief and Rs 25 billion for long-term rehabilitation. The central government provided approximately Rs 3 billion. This means the centre covered roughly 10 percent of what the state said it needed. Even if we assume the state’s estimate was inflated --- governments routinely inflate disaster demands to create negotiating room --- the gap between what was asked and what was given was not a rounding error. It was a policy choice.

To understand the scale of that choice, consider what US$4.44 billion in damage means for a state whose entire annual budget at the time was in the range of Rs 10,000-12,000 crore. The cyclone’s economic damage exceeded the state’s annual budget by a factor of three or four. It was as if someone had taken the entire productive output of the state for several years and fed it to the ocean. And the central government’s response was to cover a fraction of a fraction.

Now compare this with how the same central government responds to disasters in states that matter electorally. After the 2001 Gujarat earthquake, which killed approximately 20,000 people, the centre released Rs 3,508 crore and facilitated international aid and reconstruction on a scale that Gujarat’s political establishment could leverage for years. After the 2004 tsunami, which hit Tamil Nadu, Kerala, and Andhra Pradesh, central relief was rapid, substantial, and sustained. These were states with 26, 20, and 42 Lok Sabha seats respectively. Odisha had 21 --- and in 1999, those 21 seats were held by a combination of BJD and BJP, the same ruling alliance at the centre. Even with political alignment, the cheque was thin.

Think of it in software terms. When a system receives an input it considers high-priority --- a crash report from a production server that handles millions of users --- it triggers immediate, sustained, well-resourced response. When the same system receives an input from a staging server with minimal traffic, the response is slower, less resourced, and more likely to be deprioritized. Odisha, with its 3.5 percent of Lok Sabha seats, was the staging server. The response was calibrated accordingly.

The defence, when it is articulated at all, is that the central government’s disaster response mechanism in 1999 was underdeveloped. There was no National Disaster Management Authority (that came in 2005). There was no structured National Disaster Response Fund (the Calamity Relief Fund existed but was inadequate). The institutional architecture for large-scale disaster response simply didn’t exist in 1999 the way it does now. This defence is partially true --- the machinery was weaker. But the inadequacy of the machinery was itself a policy choice. The central government had not invested in disaster response infrastructure because states like Odisha, which needed it most, did not have the political weight to demand it.

This is the pattern that defines Delhi’s relationship with crisis-prone, electorally lightweight states: the institutional inadequacy is not the cause of poor response --- it is a symptom of the same structural neglect that produces poor response. You don’t build the fire station because you don’t think the neighbourhood matters. Then when the fire comes, you point to the absence of the fire station as the reason you couldn’t help.

Natural disasters are often described as great equalizers --- the wind doesn’t check your caste certificate before it takes your roof. This is true in the narrowest physical sense and false in every other sense that matters.

The 1999 cyclone acted as a social X-ray of Odisha’s coastal communities, revealing structures of inequality that were always present but became visible only when survival itself was at stake.

The documented evidence of caste-based discrimination in relief distribution is damning. In Jagatsinghpur district, the epicentre of devastation, relief trucks headed for Dalit villages were looted by residents of adjoining upper-caste settlements. People in Dalit hamlets were denied entry to cyclone shelters --- the very shelters that were supposed to be communal, open to all. Dalit families found themselves stranded in unsafe conditions without shelter or relief packages, while the concrete houses in elevated areas --- predominantly owned by upper and middle-caste families --- survived the surge.

This was not random cruelty. It was the spatial manifestation of centuries of caste hierarchy. Upper-caste families in coastal Odisha villages live in elevated areas and own concrete houses. Lower-caste day labourers live on the edges of villages, in low-lying areas, in kutcha structures that dissolve in heavy rain, let alone a storm surge. The geography of caste is the geography of vulnerability. When the surge came, the people living lowest drowned first. When the relief came, the people with the most social power captured it first.

ActionAid India, which mounted one of the largest NGO responses to the cyclone, specifically focused on the most affected and marginalized groups: Bengali immigrants, Dalits, women and children at risk, single women, orphans, the destitute, persons with disability, and the elderly in Erasama and Balikuda blocks. The fact that an international NGO had to specifically target these groups is itself an indictment. The state’s relief machinery, even when it functioned, reproduced the social hierarchies it was supposed to temporarily override.

A 2011 National Dalit Watch report would later document how pre-existing conditions of vulnerability and caste discrimination get “magnified and multiplied” during disasters. The cyclone didn’t create inequality. It amplified it, and then the relief distribution compounded the amplification. The Arjun Sengupta Committee Report of 2007 found that almost 88 percent of Scheduled Castes were poor and faced multidimensional vulnerability --- this was the baseline onto which the cyclone’s devastation was layered.

Here is what this means for understanding Delhi’s crisis response. When the central government allocates Rs 3 billion for a disaster affecting 13 million people, the allocation itself is inadequate. But even that inadequate allocation doesn’t reach the people who need it most. The money passes through a distribution chain that reflects existing power structures --- from Delhi to Bhubaneswar to district headquarters to block offices to village-level functionaries --- and at each node, the people with the least social power are deprioritized. The central allocation is a fraction of the need, and the fraction that reaches the most vulnerable is a fraction of that fraction.

This is not a failure of good intentions. It is a structural feature of disaster response in stratified societies. And it is a feature that Delhi’s response architecture has never been designed to address, because addressing it would require confronting caste --- a confrontation that no central government, Congress or BJP, has ever been willing to undertake within the framework of disaster relief.

The KBK Story: From Starvation to Surplus --- and Then Abandonment

Four years before the cyclone, in 1995, a different kind of crisis in a different part of Odisha forced Delhi to pay attention. But this time, the crisis was not a sudden catastrophe. It was chronic, grinding, and so old that it had become part of the national vocabulary for failure.

Kalahandi, Bolangir, and Koraput --- the KBK districts --- had been synonymous with hunger, poverty, and desperation since at least the mid-1980s. Kalahandi, paradoxically, was historically a rice surplus region --- the second-largest producer of rice in Odisha, an agriculturally fertile district that had sent rice to Bengal during the 1943 famine. But by the 1980s, a combination of failed irrigation infrastructure, exploitative moneylending, drought cycles, and the total absence of state capacity had produced starvation deaths in a district that grew more rice than it could eat. The rice was there. It was being exported to traders and markets elsewhere. The people who grew it starved.

This is important because it establishes the template. The KBK crisis was not about scarcity. It was about extraction. The same pattern --- local production of wealth, external capture of that wealth, local population left impoverished --- that defines Odisha’s mineral economy was playing out in its agricultural economy. The rice left Kalahandi the way iron ore leaves Keonjhar: through a supply chain that benefits everyone except the people at the source.

In 1995, Prime Minister P.V. Narasimha Rao visited Koraput and announced the KBK Yojana --- a special plan for the eight districts of the KBK region (Kalahandi, Bolangir, Koraput, Malkangiri, Nabarangpur, Rayagada, Nuapada, and Sonepur). The announced intent was transformative. The Revised Long Term Action Plan (RLTAP), formulated for the period 1998-99 to 2006-07, carried an approved outlay of Rs 6,251 crore.

The actual funding tells the real story.

During the first three years of the initial Long Term Action Plan (1995-96 to 1997-98), Odisha received a total additional central assistance of Rs 20.49 crore. That’s twenty crore against a plan that was supposed to spend thousands of crore. The LTAP “did not take off due to non-availability of sufficient funds” --- a bureaucratic euphemism for the central government announcing a plan, taking credit for the announcement, and then not funding it.

The RLTAP, which replaced the LTAP from 1998-99, performed somewhat better. Over the entire period from 1998-99 to 2010-11, the central government released Rs 1,963.95 crore. Total central aid under the KBK head from 1995-96 to 2005-06 amounted to approximately Rs 4,544 crore. The Institute of Applied Manpower Research later concluded that this funding “has not really changed the living condition of the people.”

If you think of development funding as a venture investment, here is what happened. The central government identified a market failure (extreme poverty in KBK). It announced a major investment (Rs 6,251 crore RLTAP). It then drip-fed the capital over a decade, never providing enough at any single point to create the critical mass needed for transformation. It then declared, after ten years of underfunding, that the investment hadn’t worked --- and withdrew.

In investing, this is called undercapitalization leading to failure. You commit to building a factory but only fund the foundation. When the factory doesn’t produce output, you blame the management, not the fact that you never built the walls. The central government’s KBK intervention followed this pattern precisely.

What Actually Worked --- And Who Did It

Despite the underfunding, something remarkable happened in the KBK region over the following decades. Kalahandi went from a symbol of starvation to a rice surplus district --- the second-largest producer of rice in Odisha. Bolangir became the largest grower of cotton in the state. By 2020, the state government could report that the KBK region had achieved food surplus.

How? Not primarily through the central government’s plans. The game-changer, according to officials, was the Indravati Dam project, completed in 1996, which brought irrigation to areas that had been entirely rain-dependent. The dam was a state project, funded and executed by the Odisha government. The irrigation it provided allowed farmers to move from single-season to multi-season cultivation, which is the single most important factor in converting subsistence agriculture to surplus agriculture.

After central funding under the RLTAP was scaled down post-2006-07, the Odisha government launched the Biju KBK Yojana --- a state-funded special development plan for the eight KBK districts. Initial allocation was approximately Rs 120 crore per year, focusing on the four essentials: Bijli (electricity), Sadak (roads), Pani (water), and livelihood. In 2015, Chief Minister Naveen Patnaik increased the fund to Rs 250 crore, explicitly stating that the increase was necessary because the Centre had stopped assistance for development in the KBK region.

Read that sentence again. The state government had to double its own KBK spending because the central government simply stopped funding its own programme. No announcement, no formal discontinuation --- just a gradual withdrawal of funds until the state was left to fill the gap with its own resources. This is fiscal abandonment disguised as budget reprioritization.

The trajectory is clear: Delhi announces, Delhi underfunds, Delhi withdraws. The state fills the gap. The transformation that eventually occurs --- Kalahandi’s move from starvation to surplus, the improvement in road connectivity and electrification --- is funded substantially by the state. But Delhi claims its share of credit because it initiated the programme, even though its actual contribution was a fraction of what was promised.

I believe with approximately 70 percent confidence that this pattern --- announcement-underfunding-withdrawal-state substitution --- is the single most consistent feature of Delhi’s relationship with Odisha across all policy domains. I would be wrong if someone can demonstrate a central programme for Odisha that was fully funded to its announced level and sustained for its full planned duration.

Building OSDMA: The Proof That Self-Reliance Works

If the KBK story shows what happens when Delhi’s promises evaporate, the OSDMA story shows what happens when a state decides to build its own institutions regardless of what Delhi does.

After the 1999 super cyclone killed nearly 10,000 people, the Odisha government faced a choice that most state governments never face in such stark terms. It could lobby Delhi for better disaster response infrastructure --- request more NDRF battalions, better early warning systems, more central funding for shelters. Or it could build its own.

It chose to build its own.

In 2000, barely a year after the cyclone, Odisha established the Odisha State Disaster Management Authority --- the first state-level disaster management body in India. This was six years before the central government established the National Disaster Management Authority in 2005. Think about what this means. Odisha, one of India’s poorest states, created a disaster management institution half a decade before the national government decided the country needed one. The state that had just lost 10,000 people to institutional failure responded by building institutions. Delhi, which had watched 10,000 people die and sent a cheque for a fraction of the damage, responded by doing nothing for six years.

What OSDMA Built

The transformation in physical infrastructure is measurable. In 1999, there were 75 cyclone shelters along Odisha’s entire coastline --- a coast that stretches 480 kilometres. Today, there are roughly 800 multipurpose cyclone and flood shelters, designed in collaboration with IIT Kharagpur to withstand wind speeds of 300 km/h and moderate earthquakes. Each shelter serves as a community centre during normal times and a life-saving refuge during cyclones.

OSDMA created India’s first end-to-end early warning system reaching the last mile. Nearly 1,200 coastal villages across all coastal districts now receive cyclone or tsunami warnings through sirens and mass messaging systems. Watchtowers in over 120 coastal locations form the backbone of the warning network. The system was designed not just to generate alerts --- India Meteorological Department was already doing that --- but to ensure that the alert reaches the fisherman in the last village on the coast, in a language he understands, with enough lead time for him to evacuate.

The evacuation protocols represent the most operationally significant innovation. OSDMA developed a mass evacuation system that coordinates state, district, and block-level authorities with community leaders, SHG networks (Mission Shakti groups played a role in local mobilization), pre-positioned relief materials, and rehearsed evacuation routes. The system is rehearsed regularly. Mock drills are conducted in coastal villages. The capacity is not theoretical --- it is practiced, tested, and refined after every cyclone.

The Results: A 200-Fold Reduction

Here are the numbers that tell the story of what institutional investment produces.

1999 Super Cyclone: Wind speeds exceeding 260 km/h. Approximately 10,000 dead. 13 million affected. Fatality rate: approximately 779 per million affected.

2013 Cyclone Phailin: Comparable intensity. Close to 1 million people evacuated in less than 48 hours. 38 dead. Fatality rate: approximately 3.6 per million affected.

2019 Cyclone Fani: Wind speeds of 215 km/h. The state evacuated 1.2 million people. Some estimates say 1.5 million were shifted in a single day. 64 dead. 43,000 volunteers deployed. 2.6 million text messages sent. 7,000 kitchens operated. 9,000 storm shelters opened. Fatality rate: approximately 3.8 per million affected.

The reduction in fatality rate from 779 to under 4 per million affected --- a roughly 200-fold improvement --- is one of the most dramatic governance transformations documented anywhere in the world. The World Bank has called it a model. The United Nations has cited it. The World Economic Forum has featured it. UNESCAP published a detailed analysis of what it called Odisha’s “zero casualty” model.

And here is the critical point for understanding Delhi’s Odisha: OSDMA was built primarily with state resources, state initiative, and state institutional capacity. The World Bank’s National Cyclone Risk Mitigation Project provided support --- US$255 million for cyclone shelters and evacuation roads in Phase I, shared between Odisha and Andhra Pradesh. But the institutional design, the evacuation protocols, the early warning system, the community-level mobilization infrastructure --- these were Odisha’s creation.

The World Bank’s own assessment credited the Odisha State Disaster Management Authority and the Government of Odisha for “their unwavering commitment to disaster preparedness and risk mitigation.” Note what is not credited: the Government of India.

What OSDMA Proves

OSDMA is the single strongest piece of evidence for a proposition that runs through every chapter of this series: when it comes to Odisha’s development, the state has been its own best resource.

Delhi sent a cheque for Rs 3 billion after 10,000 people died. Odisha built a system that saves tens of thousands of lives every cyclone season. Delhi created NDMA in 2005, six years after the catastrophe. Odisha created OSDMA in 2000, one year after. Delhi’s disaster response follows media cycles --- cameras arrive with the cyclone, aid arrives after the cameras, both leave when the next news cycle begins. Odisha’s disaster response is institutional --- it operates continuously, independent of media attention or political cycles.

If you think of governance as software architecture, Delhi’s disaster response for Odisha was reactive --- event-triggered, with no persistent state, no background processes, no caching of previous responses. OSDMA is proactive --- a continuously running system with persistent state, background monitoring, cached evacuation routes, and pre-compiled response protocols. The difference in outcomes is not surprising. Reactive systems fail under load. Proactive systems are designed for it.

The OSDMA success also reveals something uncomfortable about the “double engine” narrative. If Odisha’s greatest governance achievement --- the transformation from 10,000 deaths to near-zero casualties --- was built entirely by the state government without meaningful central government involvement, then what exactly is the “engine” that Delhi provides? The disaster management story suggests that the state engine is the one that does the actual work. Delhi’s engine provides fuel when it feels like it and takes credit when the car arrives.

The Pattern: Crisis, Camera, Cheque, Withdrawal

Step back from the specific events --- the 1999 cyclone, the KBK programme, OSDMA --- and a pattern emerges that is worth naming directly, because it repeats with such regularity that it constitutes a structural feature of Delhi’s relationship with Odisha, not an unfortunate series of coincidences.

The pattern has five stages.

Stage 1: Crisis. Something catastrophic happens in Odisha --- a super cyclone, a famine, a mining scandal, a displacement crisis. The event is severe enough to breach the threshold of national attention.

Stage 2: Media coverage. National cameras arrive. The images are compelling --- devastated villages, crying children, heroic rescuers. For a brief window, Odisha exists in the national consciousness.

Stage 3: Central response. The Prime Minister announces a relief package. A High Powered Task Force is constituted. Special allocations are made. The response is calibrated to the media attention, not to the structural need. The Rs 3 billion for the 1999 cyclone was enough to generate a headline. It was not enough to rebuild.

Stage 4: Withdrawal. The cameras leave. The news cycle moves on. The special allocation is disbursed, partially, over years. The follow-through is weak. The institutional investment that would prevent the next crisis is not made. In the KBK case, the RLTAP was funded at roughly one-third of its planned level, and then the funding dried up entirely after 2006-07.

Stage 5: State self-reliance. Odisha fills the gap. The Biju KBK Yojana replaces the central programme. OSDMA replaces the absent national disaster infrastructure. The state’s own resources --- mining royalties, fiscal discipline, institutional capacity --- are deployed to solve problems that the centre announced it would solve and then didn’t.

This pattern is not unique to Odisha. It is how Delhi relates to many peripheral states --- the northeast, the tribal belts of central India, the drought-prone regions of western Rajasthan. But in Odisha’s case, the pattern is particularly sharp because the state has both extreme vulnerability (cyclone coast, mineral belt, tribal displacement) and extreme fiscal capacity (mining revenues, budget surpluses, low debt ratios). Odisha can afford to substitute for Delhi’s withdrawal. Many other states cannot.

The political economy of this pattern is worth examining. Delhi’s response follows a logic that is rational from Delhi’s perspective, even though it is devastating from Odisha’s. Disaster relief generates immediate political capital: the PM visiting, the cheque being handed over, the high-powered committee being constituted. Long-term institutional investment generates no immediate political capital --- there is no photo opportunity in building the 400th cyclone shelter. The media covers the crisis, not the prevention. The central government, which operates on a five-year electoral cycle and a twenty-four-hour news cycle, optimizes for what the media covers. The state government, which lives with the consequences, optimizes for what actually saves lives.

This divergence between what generates political capital at the centre and what generates human capital at the state is the fundamental mismatch in India’s disaster governance. It is not a bug. It is the operating system.

Cyclone Fani, 2019: The Vindication

On May 3, 2019, Cyclone Fani --- an Extremely Severe Cyclonic Storm with wind speeds of 215 km/h --- made landfall near Puri, one of Odisha’s most sacred and densely populated cities. Fani was the strongest cyclone to hit the state in twenty years, and the strongest to hit during the summer season (most Bay of Bengal cyclones form in the post-monsoon period of October-November).

What happened next was the vindication of two decades of institutional investment.

In the 48 hours before landfall, the Odisha government executed one of the largest peacetime evacuations in human history. Approximately 1.2 million people were moved from vulnerable coastal areas to higher ground and into cyclone shelters. Some estimates put the number at 1.5 million shifted in a single day. The state deployed 43,000 volunteers, operated 7,000 kitchens to feed evacuees, opened 9,000 storm shelters, and sent 2.6 million text messages in addition to using television, sirens, and public-address systems.

The death toll: 64 people. In a storm that hit a coast with a population density of over 500 people per square kilometre. In a storm that caused over Rs 24,000 crore in economic damage.

Compare this with 1999: similar wind speeds, similar geography, similar population exposure. Ten thousand dead in 1999. Sixty-four dead in 2019. The variable that changed was not the cyclone. It was the institution.

Fani was Odisha’s proof of concept. Every rupee spent on OSDMA, every cyclone shelter built, every early warning siren installed, every mock evacuation drill conducted in coastal villages --- all of it was justified in the 48 hours before Fani made landfall. The system worked. Not perfectly --- 64 people still died, and the economic damage was staggering. But the system worked in the way that matters most: it kept almost everyone alive.

The central government’s role in the Fani response was, by all accounts, smoother than in 1999. Twenty years of institutional development, including the creation of NDMA and NDRF, had improved coordination. But the operational backbone --- the evacuation, the shelters, the community-level mobilization --- was OSDMA’s. The NDRF provided specialist teams. The Indian Armed Forces assisted. But the million-person evacuation was run by Odisha’s own machinery.

Even in the disaster management space, however, the old patterns of discrimination persisted. Investigations after Fani documented how Dalit communities in the affected areas experienced delays in relief distribution and caste-based discrimination in access to shelters and relief materials. A 2020 study published in the Indian Journal of Social Work found evidence of prejudice in the distribution of relief during Cyclone Fani, with lower-caste communities systematically receiving less and later than their upper-caste neighbours.

I should be honest about what this means for the argument. OSDMA’s achievement in reducing fatalities is unambiguous and extraordinary. But the social distribution of vulnerability and the social distribution of relief remain deeply unequal. Odisha has solved the engineering problem --- how to evacuate a million people in 48 hours. It has not solved the social problem --- how to ensure that evacuation and relief reach all communities equally. The engineering problem was solvable through institutional design. The social problem requires confronting caste, which is a much harder engineering project.

The Deeper Argument: What Disaster Response Reveals About the Federal Bargain

Here is the uncomfortable truth that the cyclone-KBK-OSDMA triptych reveals about Delhi’s relationship with Odisha.

Delhi’s crisis response is not designed to address structural vulnerability. It is designed to manage the political optics of crisis. The difference is fundamental.

A system designed to address structural vulnerability would invest in prevention before the crisis. It would fund cyclone shelters before the cyclone. It would fund the KBK programme at its planned level before the starvation deaths. It would build disaster management institutions before the disaster. It would treat the recurring pattern of crisis in Odisha --- cyclones every few years, chronic poverty in the tribal belts, displacement from mining --- as a systemic problem requiring systemic investment.

A system designed to manage political optics does something different. It responds after the crisis. It calibrates its response to media attention. It announces large numbers that sound impressive in press conferences but are fractions of the actual need. It constitutes high-powered committees that meet a few times and then dissolve. It moves on to the next crisis in the next state that happens to be in the news cycle.

Delhi’s system is the second kind. This is not because the individuals involved are callous. It is because the incentive structure of Indian federalism optimises for visibility over efficacy. The Prime Minister visiting a cyclone-hit area generates more political return than the Prime Minister approving a ten-year cyclone shelter construction programme. The announcement of Rs 6,251 crore for KBK generates more coverage than the actual disbursement of Rs 1,963 crore over thirteen years. The system is optimised for announcements, not outcomes.

Odisha’s achievement with OSDMA is the mirror image. It represents a state government that optimised for outcomes over announcements. There was no grand national launch of OSDMA. There were no crore-denominated press conferences. There was a sustained, methodical, unglamorous process of building shelters, installing sirens, rehearsing evacuations, and learning from each cyclone. The outcomes --- a 200-fold reduction in cyclone fatalities --- speak for themselves, but they speak quietly, because the process that produced them was designed for efficacy, not visibility.

This contrast --- Delhi’s visibility-optimised response versus Odisha’s efficacy-optimised institution-building --- is perhaps the clearest illustration of why the centre-state relationship in India produces the outcomes it does. The centre operates on a logic of political return per rupee spent. The state operates on a logic of lives saved per rupee spent. When these logics diverge, the centre will always underinvest in states that don’t generate sufficient political return --- which means states with few Lok Sabha seats, weak media presence, and no leverage over government formation.

Odisha is such a state. Its 21 Lok Sabha seats, its absence from the national media cycle except during disasters, its lack of representation in senior Union Cabinet positions --- all of this makes it structurally invisible to a centre that allocates attention, resources, and institutional investment based on political arithmetic.

The cyclone and the cheque are the perfect metaphor for this relationship. The cyclone is real, devastating, and recurring. The cheque is real, insufficient, and one-time. What bridges the gap between the cyclone’s destructive power and the cheque’s inadequacy is the state’s own capacity --- built with the state’s own resources, sustained by the state’s own institutional discipline.

This is not a story of heroic self-reliance. It is a story of forced self-reliance. Odisha did not choose to build OSDMA instead of receiving adequate central support. It built OSDMA because adequate central support was never going to come. The self-reliance is admirable. The fact that it was necessary is the indictment.

Sources

“1999 Odisha cyclone.” Wikipedia. https://en.wikipedia.org/wiki/1999_Odisha_cyclone --- Death toll, damage estimates, central government allocation, and relief gap data.
“Odisha Super Cyclone 1999.” ActionAid India. https://www.actionaidindia.org/emergency/odisha-super-cyclone/ --- Documentation of relief work focused on Dalits, Bengali immigrants, women, and marginalized communities.
“25 years of super cyclone: For Odisha’s coastal villages, the 1999 tragedy’s memories are still alive, but so is their resilient spirit.” Down to Earth, October 2024. https://www.downtoearth.org.in/natural-disasters/25-years-of-super-cyclone-for-odishas-coastal-villages-the-1999-tragedys-memories-are-still-alive-but-so-is-their-resilient-spirit
“Landless Dalits Hit Hardest By Disasters Are Last To Get Relief.” LSE South Asia Blog, November 2019. https://blogs.lse.ac.uk/southasia/2019/11/01/long-read-landless-dalits-hit-hardest-by-disasters-are-last-to-get-relief/ --- Caste-based discrimination in disaster relief in Odisha.
“Delay, Despair And Discrimination Against Dalits Even After Cyclone Fani.” Outlook India, 2019. https://www.outlookindia.com/national/india-news-delay-discrimination-and-despair-challenges-after-a-cyclone-news-330399
“Discrimination during pandemic, cyclones puts lives of dalits at risk.” Down to Earth, 2021. https://www.downtoearth.org.in/governance/discrimination-during-pandemic-cyclones-puts-lives-of-dalits-at-risk-78580
“Prejudice In Times Of Natural Disasters: An Assessment Of Caste Discrimination During Cyclone Fani In Odisha.” ResearchGate, 2023. https://www.researchgate.net/publication/374419263_Prejudice_In_Times_Of_Natural_Disasters_An_Assessment_Of_Caste_Discrimination_During_Cyclone_Fani_In_Odisha
“Reminiscing KBK Plan: Myth and Reality of Development in India.” Society for the Study of Peace and Conflict. https://www.sspconline.org/opinion-analysis/reminiscing-kbk-plan-myth-and-reality-development-india --- KBK funding gap analysis.
“KBK Districts: A unique approach towards development.” Government of India, KBK Programme. https://kbk.nic.in/pdf/KBKProfile.pdf --- Official RLTAP data and expenditure figures.
“Monitoring of Special Plan for KBK Districts.” Planning & Convergence Department, Government of Odisha. https://pc.odisha.gov.in/schemes/monitoring-special-plan-kbk-districts
“Biju KBK fund increased to Rs 250 crore.” Daily Pioneer, 2015. https://www.dailypioneer.com/2015/state-editions/biju-kbk-fund-increased-to-rs250-crore.html --- State-funded KBK programme after central withdrawal.
“Hunger deaths to ‘rice bowl’: How Odisha’s Kalahandi-Balangir-Koraput corridor turned a corner.” ThePrint, 2021. https://theprint.in/india/hunger-deaths-to-rice-bowl-how-odishas-kalahandi-balangir-koraput-corridor-turned-a-corner/784677/
“KBK Region in Odisha Needs a New Development Roadmap.” OdishaPlus, February 2026. https://odisha.plus/2026/02/kbk-region-odisha-new-development-roadmap/
“Rs 2537.50 crore released so far for development of KBK region.” Press Information Bureau, Government of India. https://www.pib.gov.in/newsite/PrintRelease.aspx?relid=113737
“Biju KBK Plan: Modalities and Guidelines.” Odisha Review, November 2006. https://magazines.odisha.gov.in/orissareview/nov-2006/engpdf/132-136.pdf
“Odisha’s turnaround in disaster management has lessons for the world.” World Bank, November 2023. https://www.worldbank.org/en/news/opinion/2023/11/03/odisha-s-turnaround-in-disaster-management-has-lessons-for-the-world
“From storm to strength: Odisha’s zero casualty model for community-centered disaster resilience.” UNESCAP, 2024. https://www.unescap.org/blog/storm-strength-odishas-zero-casualty-model-community-centered-disaster-resilience
“Cyclone Devastation Averted: India Weathers Phailin.” World Bank, October 2013. https://www.worldbank.org/en/news/feature/2013/10/17/india-cyclone-phailin-destruction-preparation
“India Averts Devastation from Cyclone Phailin.” World Bank, April 2014. https://www.worldbank.org/en/results/2014/04/10/india-averts-cyclone-phailin-devastation
“World Bank Board Approves $153 Million Support to Odisha Disaster Recovery Project.” World Bank, February 2014. https://www.worldbank.org/en/news/press-release/2014/02/21/world-bank-approves-support-odisha-disaster-recovery-project
“Lessons in disaster relief from the world’s most cyclone-battered state.” World Economic Forum, December 2019. https://www.weforum.org/stories/2019/12/disaster-relief-lessons-cyclone-fani-odisha/
“Cyclone Fani.” Wikipedia. https://en.wikipedia.org/wiki/Cyclone_Fani --- Evacuation numbers, casualties, and response details.
“Cyclone Fani response in Odisha.” ReliefWeb, 2019. https://reliefweb.int/report/india/cyclone-fani-response-odisha
“Odisha State Disaster Management Authority (OSDMA): A Benchmark for Disaster Preparedness and Management in India.” International Journal of Research and Innovation in Social Science. https://rsisinternational.org/journals/ijriss/articles/odisha-state-disaster-management-authority-osdma-a-benchmark-for-disaster-preparedness-and-management-in-india/
“NITI MPI Shows Odisha’s KBK Syndrome Lingers On From Narasimha Rao To Naveen Patnaik.” OdishaTV. https://odishatv.in/news/exclusive/niti-mpi-shows-odisha-s-kbk-syndrome-lingers-on-from-narashima-rao-to-naveen-patnaik—164926
CAG Performance Audit Report on KBK Districts. Comptroller and Auditor General of India. https://saiindia.gov.in/uploads/old_reports/state/Orissa/2007/KBK_Districts/cv_pa_overview.pdf

Source Research

The raw research that informs this series.