India crossed 20 percent ethanol blending in 2025, five years ahead of the deadline it had already advanced once. By April 2026, E20 was the default fuel at pumps nationwide. Measured against its own target, the ethanol-blended petrol programme can be seen as one of the most successfully executed industrial policies of the past decade. That is precisely why it now deserves harder scrutiny.
The question is not only whether E20 damages engines or reduces mileage. It is who pays for the blend, who collects the rent, and whether the farmer—the programme’s principal moral justification—is anywhere near the front of that queue.
The government’s ledger is impressive on its face. Since 2014-15, the ministry of petroleum and natural gas puts foreign exchange savings at over ₹1.97 lakh crore, crude substitution at 316 lakh metric tonnes, avoided CO₂ emissions at 952 lakh metric tonnes and payments into the farm economy at ₹1.66 lakh crore. Nor is the energy-security argument merely rhetorical.
When the West Asia conflict pushed crude up by 70-80 percent after February 2026, ethanol continued to arrive at a fixed, administered price of roughly ₹70 a litre while the Indian basket touched nearly $135 a barrel.
Domestic pump prices rose 7-8 percent. Oil marketing companies absorbed an average under-recovery of about ₹11 a litre—some ₹21,300 crore between March and June—and petrol that might have cost ₹125 was sold at around ₹95 in Delhi. A fifth of the petrol pool priced in rupees, immune to the dollar and to the Hormuz Strait, is a genuine hedge. Any honest critique must concede this.
But a hedge is insurance, and insurance has a premium. The programme’s central evasion is that this premium has never been priced, disclosed or fairly allocated.
Begin with cost. In a written reply to the Lok Sabha in July, the government disclosed that the weighted average ex-mill price of ethanol for the current supply year is ₹66.61 a litre, and that oil marketing companies’ delivered cost, inclusive of GST and freight, hovered at ₹71.10-71.21. Maize-based ethanol, now the single largest feedstock, is procured at ₹71.86 ex-mill.
At any normal crude price, this is comfortably above the refinery cost of the petrol it displaces. Blending, in other words, is no longer a cost-reducing activity; it is a cost-plus one that happens to look cheap during a war. The consumer bears this twice over.
Ethanol carries roughly a third less energy per litre than petrol. The petroleum ministry itself concedes a 3-5 percent fuel-economy loss. Niti Aayog’s 2021 roadmap was more candid, projecting 6-7 percent for four-wheelers designed for unblended petrol and calibrated for E10, and 3-4 percent for comparable two-wheelers. Yet E20 is sold at exactly the price of the fuel it replaced.
A less energetic fuel sold at parity is a transfer, and it flows from the two-wheeler owner to the ethanol value chain. That transfer could be neutralised at a stroke, and Niti Aayog said as much in 2021 when it recommended that blended fuel be priced lower, or that tax relief compensate for lost efficiency. Neither happened.
The excise exemptions extended in June 2026 to E22, E25, E27 and E30 are, by the government’s own explanation, only a device to prevent double taxation—not a rebate to the driver. To this day, the ministry has published no per-litre cost stack for E20: not the depot cost of petrol, not the blending cost, not the tax on each component. Without that disclosure, the claim that E20 is dearer to make cannot be tested, and the question of who pockets the difference cannot be answered.
The distribution claim is weaker still. The ₹1.66 lakh crore is not paid to farmers; it is paid to distilleries and mills, from whom it must travel down to the farmgate. Whether it does is an empirical question with an uncomfortable answer. Agriculture ministry data show that the average mandi price of maize stayed below the minimum support price in every single month from March 2025 to August 2026.
In August, maize averaged just over ₹2,008 a quintal against an MSP of ₹2,410—for a crop the Commission for Agricultural Costs and Prices sets at ₹1,544. This is the crop farmers were urged to expand for ethanol; kharif acreage rose to about 9.5 million hectares. Demand did not follow, partly because the state undercut it: surplus Food Corporation of India rice was made available to distilleries at ₹2,320 a quintal, reportedly far below FCI’s own acquisition cost, while maize import restrictions were eased.
In ethanol supply year 2025-26, grain feedstock accounts for roughly 72 percent of allocation against 28 percent from sugarcane. Karnataka’s distillers told the state government plainly that with output caps and thin margins they could not buy maize at MSP.
Cane tells a parallel story. Fair and remunerative price was raised by ₹10 to ₹365 a quintal for 2026-27; maize MSP by ₹10. Of ₹1,12,740 crore in cane dues for 2025-26, about ₹99,961 crore had been paid by April 2026—leaving nearly ₹13,000 crore outstanding, with farmer bodies putting the arrears at ₹16,087 crore.
The draft Sugarcane (Control) Order, 2026 formally recognises ethanol as a mill output, counting 600 litres as one tonne of sugar, yet still anchors the grower’s entitlement to sugar alone. The rent, predictably, settles where the administered price is fixed: at the distillery gate, cushioned by 5 percent GST, interest subvention and soft loans that pulses and oilseeds never received.
Then there is the resource account. Rice-based ethanol carries a water footprint estimated by the Food Secretary at about 10,790 litres per litre, maize 4,670 and sugarcane 3,630. Industry correctly notes that distilleries themselves consume only 3-4 litres; that is the point—the water is spent in the field, in Maharashtra, Punjab and Madhya Pradesh, where aquifers are already failing.
Meanwhile, installed capacity has run ahead of demand: about 1,900 crore litres against roughly 1,048 crore litres contracted, leaving over 450 crore litres idle. Stranded capital of that scale does not sit quietly; it lobbies for E30, for exports, for more grain. None of this argues for retreat to E0.
It argues for five corrections. Publish the full cost stack of a litre of E20 and audit the programme’s net benefit annually. Use the excise lever to make the motorist’s energy-adjusted cost neutral, as Niti Aayog advised. Replace feedstock-wise administered pricing with a formula linked to petrol trade parity plus an explicit, budgeted subsidy, so the transfer is visible in the Finance Bill rather than buried in the pump price.
Stop subsidising FCI rice into distilleries while telling farmers to plant maize, and tie ethanol allocations to verified MSP-linked procurement. And legislate revenue-sharing so that ethanol realisations reach the cane grower.
Above all, do not proceed to E30 before E20 has been independently audited. A policy that is right in its energy logic can still be wrong in its distribution. On the only test that matters for a farm-fuel programme—what happens at the farmgate—E20 has not yet passed.
Santhosh Kumar P K | Professor & Director, Centre for Budget Studies, Cochin University of Science and Technology
(Views are personal)
