How the Lights Went Out

The Anatomy of a Failure

The lights went out on April 25, 2026. At 10:30 PM the Kerala grid collapsed into a 438-megawatt deficit. A month later the LDF government lost the election. On June 4, the incoming UDF administration under Chief Minister V.D. Satheesan tabled a White Paper on the finances it had inherited.

Three failures produced the blackout. A forty-year refusal to build nuclear, large hydro, or private thermal at home. A 2023 regulatory ruling that voided 465 megawatts of cheap long-term power. A procurement desk that let 400 megawatts of summer swaps lapse unrenewed on April 21. The White Paper put a number on what they cost.

I. The Hardware and the Ledger

An electricity system has three physical layers and one financial layer. Generation manufactures the power. Transmission carries the voltage across distance. Distribution is the last mile of wire. KSEB runs the wires competently; state power losses sit at 7.55 percent, below the national average. The hardware is dated. There is still no modern smart-meter network, the rollout stalled for years to protect union jobs, and billing depends on a meter reader at the gate rather than real-time data. The crisis of April 2026 lived in the fourth layer. Procurement is the desk that decides who buys what power, on what contract, at what price.

Demand forms a curve. It crests on summer evenings, when air conditioners and lights switch on together, and bottoms out around 4 AM. Different plants fit different parts of the curve. Coal and large hydro supply a steady, cheap base through the day. Gas turbines and battery storage are sprinters, available the instant the evening peak hits. Solar peaks at noon and is gone by dusk. Kerala's demand crests between 7 and 11 PM, after the solar has vanished.

KSEB bridges the gap with contracts. Long-term Power Purchase Agreements run 25 years at a fixed rate and insure against the open market. Short-term power trades on the Indian Energy Exchange, where the price can move from ₹2 to ₹20 a unit inside a week. Swap deals are barter: Kerala draws power from Madhya Pradesh through the summer and returns it during the monsoon, when its own dams are full and Madhya Pradesh needs topping up. No money changes hands. The arrangement asks only that someone keep track of the dates.

II. The Policy of Refusal

Private enterprise built this grid. In 1900 a British tea company commissioned the state's first hydroelectric plant at Pallivasal. Kannan Devan Hills Plantations still holds a legacy distribution licence in Munnar; Thrissur Corporation runs another historical carve-out. KSEB tolerates these antiques and admits nothing new beside them.

Kerala generates barely a quarter of the power it uses and imports the rest. The shortfall follows a political template rather than geography. From the 1940s to the 1980s the state poured concrete: Pallivasal, Sabarigiri, Idukki. Sabarigiri carried Kerala into a power surplus in 1966. Then the building stopped.

The Silent Valley campaign of the 1970s halted a dam to save an intact rainforest, and won on the merits. The victory hardened into a method. The state turned it against every later proposal regardless of the specifics. Athirappilly, floated in 1979, was buried in 2021. Pooyamkutty went the same way.

Kerala has no thermal coal, no natural gas, and a settled aversion to nuclear power. Declining to build heavy generation inside the state is a defensible choice, but only if something replaces it. Nothing did. The state left pumped-storage hydro unbuilt, blocked the smart meters a modern grid needs, and priced private industrial power out of reach. Imports filled the space where engineering should have been.

III. The Autopsy and the Verdict

On April 14, 2026, daily consumption reached 112 million units and peak demand touched 6,012 megawatts. Both were records.

On April 21 the 200-megawatt summer swap with Madhya Pradesh expired. The 200-megawatt swap with Punjab lapsed the same afternoon. Four hundred megawatts left the system at once, with nothing lined up to replace them. The expiry dates had sat on the calendar at Vydyuthi Bhavanam for twelve months. The paperwork was never done.

By April 25 the shortage stood at 438 megawatts.

The lapsed swaps were the proximate cause. The structural one had been settled in a courtroom. In December 2014 the UDF ministry signed long-term PPAs for 865 megawatts at ₹4.26 a unit, fixed for 25 years. The power flowed cleanly for seven years. In 2023 the state regulatory commission cancelled 465 megawatts of those contracts on a procedural technicality, and the LDF ministry accepted the ruling without contest. Power that had cost ₹4.26 a unit now costs ₹10 on the spot exchange. The daily premium for replacing it runs between ₹15 and ₹20 crore.

On June 4, 2026, Satheesan's White Paper named KSEB among the primary anchors dragging the exchequer down. It put the utility's accumulated revenue gap at ₹6,645 crore and itemised the premiums paid to import 82 percent of the state's power at the height of the crisis.

IV. The Preventable Deficit

The heatwave was real. The state ran about 4 degrees above the seasonal norm, and a weak monsoon had left the reservoirs low. The deficit still did not require the weather. Had the 2014 PPAs survived, 465 megawatts of baseload would have flowed and the 438-megawatt shortage would not have existed. Had the procurement desk renewed the 400 megawatts of swaps, the gap would have stayed closed. Renewing a swap takes a signature, not a power plant. Pumped-storage hydro would have stored cheap solar at noon and released it at 10 PM. The topography is there. The concrete is not.

V. The Nuclear Calculus

Kerala refuses nuclear power and frames the refusal as ecology. No government has published a paper weighing a local reactor against the imported coal the state burns instead. The capability is not hypothetical. Kudankulam, in Tirunelveli, has run two 1,000-megawatt reactors without incident since 2013. Kaiga, in Karnataka, runs four units. Kerala already draws Kudankulam's output through the southern grid every day. It takes the electricity and refuses the reactor.

India is now developing Small Modular Reactors, which need a fraction of the land of the older plants and no coastline. The objections Kerala fixed in the 1980s describe a technology the industry has moved past. The Silent Valley logic, sound against a dam through a rainforest, has been applied to every reactor proposal for forty years. The choice it forecloses is concrete: a modern reactor in Kasaragod against a failing grid in Malabar, propped up by a ₹6,645 crore revenue gap. The two outcomes have never been set side by side.

VI. The Unfailing Monopoly

In Tamil Nadu, auto plants buy electricity directly from private producers such as Adani Green. In Gujarat, Torrent Power competes with the state utilities for industrial customers. When a state utility mishandles procurement there, large consumers arrange their own supply. In Kerala they cannot.

The monopoly survives because CITU runs the workforce. CPI(M)'s union wing has organised KSEB's permanent staff for most of fifty years. KSEB employs 28,000 people and carries the highest staff-per-unit ratio of any large utility in the country. A private generator selling into an industrial park needs about thirty. Every megawatt bought from a private supplier is a megawatt routed around a CITU substation.

The question now is whether the UDF can dismantle the arrangement when its own labour wing benefits from it. INTUC, the Congress affiliate, runs its own bodies inside KSEB and lives off the same headcount; the maths is the same whichever colour the union flag carries. The Congress in Kerala has historically chosen to compete with CITU for the membership rather than retire the institution that produces it. The Chandy ministry held office from 2011 to 2016. It signed the cheap PPAs the LDF later cancelled. It did not touch the staffing structure that surrounds them.

The 2025 net-metering draft showed the reflex plainly. KSEB proposed paying rooftop solar generators ₹2 to ₹3 a unit while charging them ₹8 at retail. A household spending ₹4 lakh on panels would never recover the cost. Consumer groups forced a revision, but the first draft had already stated the instinct: a unit generated on a roof is a unit KSEB cannot bill.

VII. The Exported Dilemma

Germany spent €200 billion on its energy transition and shut its last nuclear reactor in 2023. On cold, windless winter nights its grid now runs on domestic coal and imported power from Polish coal plants. The carbon was not removed. It was relocated.

Kerala runs the same trade on a smaller income. Its policy rests on hydro and solar; the grid does not read policy. When demand reaches 6,135 megawatts, the balance arrives from the coalfields of Chhattisgarh.

The state moves electricity the way it moves labour. The software engineer leaves for Bengaluru, the nurse for the Gulf; the scaffolding worker comes in from Murshidabad. Power follows the same route. The heavy generation happens in Korba, the electricity returns on high-tension lines, and the environmental credit stays in Thiruvananthapuram. Both arrangements hand the hard part of a modern economy to people with the least political voice.

VIII. The Ledger of a Decade

The decade was not all failure. The distribution network held; sub-8 percent losses are a genuine achievement. What the system would not do was modernise, and the front office let the contracts lapse. The 2023 PPA cancellation is the most expensive financial decision in the state's modern power history. The swap lapse was administrative negligence. Smart tariffs are still a talking point twelve years after they were proposed.

None of this is partisan. Both fronts halted hydro, avoided storage, and leaned on imports. The UDF signed the cheap PPAs; the LDF let them be cancelled. The UDF now inherits the ledger it helped write.

IX. The Inherited Agenda

April 2027 is eleven months away. The agenda already sits on the desk at Pattom.

Sign the long-term PPAs

The previous government scrapped a private bid at ₹5.80 a unit; revive it. Fixed power at ₹5.80 beats spot power at ₹10, and every day the contract waits, the exchange collects the difference.

Renew the swaps before the heat returns

The 400 megawatts from Madhya Pradesh and Punjab have to be back on the books before the monsoon recedes.

Open the books on the 2023 cancellation

The government has the authority to run a post-mortem on the discarded ₹4.26 PPAs. If the failure was procedural, rewrite the rulebook so a regulatory commission cannot cut cheap power from the grid without a costed alternative on the table.

Build the pumped storage

Existing reservoirs can hold midday solar for the night peak; it is the cleanest engineering fix available. Tender the first pour this fiscal.

Make the monopoly compete

Lower the charges KSEB levies on private producers so they can bid for industrial clients in places like Kanjikode. The utility's captive revenue will shrink and its permanent headcount will fall. That is the efficiency dividend, and it belongs to the consumer.

Settle the nuclear question on paper

Direct the State Planning Board to publish a cost-benefit analysis setting the price of imported coal against the operating record of India's modern reactors. The exercise commits no concrete. It commits the state to reading the arithmetic of its own grid.

If not imported coal, then what

Kerala has dodged the question for forty years. It rejects nuclear, rejects private thermal, rejects large dams, and penalises rooftop solar. It refuses every form of generation it could build, and never refuses the electricity.

The state has treated the grid as a weather event for two summers running. April 2026 was the interest on decades of outsourced decisions. April 2027 is coming, the thermometer will climb, and the meter does not care who tabled the White Paper.