The economics of energy independence Last financial year India imported 88 per cent of its crude oil, at a cost of $137 billion — close to what it spends on defence.
Deficiency one: the solar cell Solar is the most advanced part of the story, and therefore the most revealing.
India's main solar research centre, at IIT Bombay, has received about Rs 200 crore over 15 years — roughly $23 million, a sum India's solar deployment programmes get through in a fortnight.
But no home-grown cell technology has yet crossed into commercial production.
A factory making licensed cells on imported machines meets the rules just as well as one making cells India actually invented.
Replacing imported oil with home-made electricity is India's surest defence against the next Hormuz. Whether India can do it will depend less on oil tankers than on what its factories can build.
In the spring of 2026, the United States and Iran went to war, and the Strait of Hormuz was shut for three-and-a-half months. This is the channel that carries about a fifth of the world's oil and most of India's cooking gas, and yet India kept the lights on and the stoves lit. There was no petrol rationing, and cooking-gas cylinders kept arriving. The crude basket spiked to $157 a barrel and then came back down; the rupee touched a record low and recovered. A decade of adding new suppliers meant cargoes could be re-sourced from the Americas and West Africa within weeks, the navy escorted tankers through the Gulf of Oman, and the government moved fast to shield consumers. By mid-June the strait reopened. It was a crisis handled well, and the country is right to take the credit. The crisis passed. The bigger question is whether India would cope as well if Hormuz stayed shut for six months or a year instead of three-and-a-half. The answer is probably not. Beneath the calm surface, the country's margin for error was thinner than it appeared. So the smart thing to do now, while the pressure is off, is to build a kind of resilience that does not depend on the next crisis being short. That means looking past the usual tools — bigger reserves, more suppliers, a stronger navy, all of which matter — to a less obvious one: the factory floor. The lasting way out of the Hormuz trap is not to buy the same oil and gas from somewhere else. It is to need less of it. India should replace imported fuel with domestic electricity wherever it can, which means electrifying transport and cooking and building the solar, wind, grid and storage to power them. But there is a catch. Replacing imported fuel with domestic electricity only works if India can build the technologies that make that transition possible. That is where four industrial gaps still stand in the way. The economics of energy independence Last financial year India imported 88 per cent of its crude oil, at a cost of $137 billion — close to what it spends on defence. Most of that oil is burned in transport, and almost none of it has to be: scooters, cars, buses and increasingly trucks can run on electricity made at home from sunlight, which need not be shipped through a contested waterway. Take cooking for example. India spends nearly $12 billion a year importing cooking gas, about 60 per cent of what it uses and roughly 90 per cent of that through Hormuz. Move a household from a gas cylinder to an induction stove and its energy shifts to the grid, which runs mostly on domestic coal and renewables — from a fuel that is 90 per cent imported to one largely made at home. These are not trivial numbers. Against a current account deficit of $23.3 billion last year, the imported gas used for household cooking is worth about $10.4 billion, so even a partial switch to domestic electricity could shave a tenth to nearly half of that deficit, before counting subsidy savings. These are illustrative calculations, not forecasts. Even so, they point clearly in one direction. The strategic case for electrification is therefore straightforward. The harder question is whether India can build the technologies it depends on. Solar, batteries and the electricity grid are all expanding rapidly, but the most valuable parts of each supply chain still lie beyond India's reach. Four technology gaps explain why. Deficiency one: the solar cell Solar is the most advanced part of the story, and therefore the most revealing. India now has the world's second-largest solar module assembly base — between 144 and 210 gigawatts of capacity, against home demand of about 40 gigawatts a year. This is a genuine achievement of the Production Linked Incentive scheme and the tariff walls and purchase rules that came with it. But a module is the easy part: a precision lamination job, sixty-odd cells sealed between glass and aluminium, on a line that can be built in 18 months. The value and the difficulty lie further up the chain, in the sequence that runs from sand to polysilicon to ingot to wafer to cell, and the higher up you go, the thinner India's capability gets. The PLI scheme awarded 48 gigawatts of cell capacity, yet by mid-2026 India had only about 27 gigawatts of actual cells.
India's capacity peaks at the module, the easiest and cheapest step, and collapses upstream where the value and the difficulty sit.
The deeper problem is not the volume but the ownership. Every high-efficiency solar cell made in India today uses technology licensed from abroad — the PERC and TOPCon designs from Chinese and Korean firms, and heterojunction from a Norwegian company Reliance bought in 2021. Solar technology also keeps moving: cell efficiency improves about every five years, and the company that owns each generation licenses it with a deliberate lag, keeping the newest version and selling the older one. So when tandem cells become commercial at the end of this decade, India's current lines will need a fresh licence, or new equipment, or both, on terms the licensor sets. The imbalance becomes starker when you look at research. India's main solar research centre, at IIT Bombay, has received about Rs 200 crore over 15 years — roughly $23 million, a sum India's solar deployment programmes get through in a fortnight. The country spends about a hundred times more putting up panels than on the research that would let it own what it puts up. The talent is there: an IIT-incubated firm has built a silicon-perovskite tandem cell at 29.8 per cent efficiency, near the global frontier. But no home-grown cell technology has yet crossed into commercial production. Deficiency two: the battery cell Batteries tell a sharper version of the same story. India imports close to 100 per cent of its lithium-ion cells, and the Rs 18,100-crore scheme meant to change that had, by late 2025, delivered just 2.8 per cent of its 50-gigawatt-hour target, from a single company. The knowledge gap here is starker: India's two largest battery firms hold seven lithium-ion patents between them, while China's CATL holds around fifty thousand. A few serious Indian efforts do exist. One firm ships cells built to a Japanese recipe using Chinese materials; another is redesigning the electrode from scratch and building its own machines. But these firms are small, short of money, and years from scale, while the gigafactories announced by the big conglomerates are mostly delayed and mostly built on Chinese or Korean technology. The pattern is exactly the one seen in solar — licences without the research to absorb and improve on them, which one battery founder called "rented capability." India can manufacture the battery, but the technology still belongs to someone else. Deficiency three: the electrical steel the grid needs Of all the technologies in this story, electrical steel is probably the easiest to overlook—and one of the hardest to replace. India is the world's second-largest steelmaker; it produced 168 million tonnes last year and became a net exporter again. Yet it still paid $8–10 billion for the special grades its factories cannot make, and the most worrying of these is the one the grid depends on. Cold-rolled grain-oriented electrical steel sits at the heart of every transformer, and India made barely 50,000 tonnes of it against demand of 400,000 tonnes, importing the rest from Japan, Korea, China and Russia at about a billion dollars a year. The Ministry of Steel told Parliament that the technology to make it "is not available with any Indian steelmaker." The related non-grain-oriented steel that goes into EV motors is worse off still, with no Indian plant making it at scale. This is the gap that most directly chokes the whole plan. You cannot run a country on electricity if you cannot build the grid to move it; you cannot build the grid without transformers; and you cannot build transformers without grain-oriented electrical steel. India is planning the biggest rise in electricity demand in its history — electric transport, electric cooking, air-conditioning for a warming country — and the single most important input to that grid is a steel it does not know how to make. The one serious project to fix this, a joint venture in Nashik, will lift home capacity to 350,000 tonnes by 2028, roughly today's demand — but by 2028, with demand growing 10 to 12 per cent a year, the need will be far higher.
Even after the Nashik plant opens in 2028, demand for grain-oriented steel has already climbed past it.
Deficiency four: the machines and the materials underneath The fourth deficiency is the one that sits beneath the other three, and it is the reason they are so hard to fix. Even where India builds a factory, the machines inside it and the raw materials feeding it usually come from abroad. Take the machines first. The equipment that makes a solar cell is made by only a handful of companies in the world, and most of them are Chinese; two Chinese toolmakers alone supplied more than half of the solar manufacturing equipment brought in by India's 10 largest solar firms. Batteries are the same. China has restricted the export of cell-making equipment to India, which is precisely why the startup redesigning its electrode also had to build its own machines. Building a factory is not the same as controlling the technology behind it. Then take the materials. At the very top of the solar chain sits polysilicon, the purified silicon from which wafers and cells are grown, and India makes almost none of it — capacity of around 3 gigawatts against an assembly base of well over a hundred. The same is true along the battery chain, where the processed cathode and anode materials, and much of the lithium itself, are imported. So even a genuinely Indian-run factory, making cells on Indian soil, can still depend on foreign machines to build the line and on foreign feedstock to run it. This is the deepest layer of the dependence, and until it is addressed, the first three deficiencies cannot truly be closed — only relocated one step upstream. Why the factories are hollow Faced with all this, it is tempting to conclude that India's industrial policy has failed and should be scrapped. That is the wrong lesson. The subsidy schemes did what they were built to do: they created manufacturing capacity fast, at the easier downstream end of each chain, and they were right to start there. Modules before cells, packs before cells, ordinary steel before special steel — that is the correct order, and a production subsidy is the correct tool for it. The problem is not that India built the factories. The problem is what the policy asked of them, and what it did not. India's schemes reward production. They pay for volume and for domestic value addition, but they do not reward technological progress: there is no rising efficiency target, no patent requirement, no push to move from licensee toward licensor, and no penalty for standing still on someone else's design. A factory making licensed cells on imported machines meets the rules just as well as one making cells India actually invented. The protection comes with no strings, and protection with no strings buys advanced assembly lines for foreign technology. That is exactly what India has today: real factories, but hollow at the core.
Almost everything India makes sits high on scale and low on ownership, the opposite corner from where Samsung and TSMC ended up.
Subsidies need a stopwatch The remedy is not more subsidy or higher tariffs. It is a condition attached to the subsidy — and here India can learn from China. The popular picture of China is of a state writing blank cheques to favoured champions, but that picture is a decade out of date. China tried unconditional subsidy in the 2000s, with flat payments per kilowatt-hour of battery capacity and credit lines with no strings, and got phantom electric buses, collapsing solar giants, and an official reckoning: in Beijing's own words, it had funded "blind investment" that bred a "dependency syndrome." What replaced it after 2016 was the opposite, with every yuan tied to real engineering. Battery subsidies were pegged to an energy-density floor that rose every 12-18 months, and if a firm missed the floor the money stopped. Solar capacity was reserved for modules that cleared efficiency targets which climbed on a schedule, and certifications, even for small firms, expired every three years and had to be re-earned. CATL and BYD were not spared this pressure; they were built by it. Meanwhile more than 400 Chinese EV startups that could not keep up were allowed to fail, with no rescue.
China tied every subsidy to a bar that rose every 12 to 18 months and expired every three years; India's support has stayed flat.
