Sugar prices rose roughly 10% in the past month. That is not just a grocery story. It is a fuel story. India's ethanol programme now draws from the same supply chain that feeds sugar mills, and when sugar gets expensive, the economics of blending shift with it. Distillers that could sell into a rising sugar market are instead locked into administered ethanol prices. The tension between these two markets is the programme's least discussed vulnerability, and it is live right now.
India's ethanol blending programme hit a 2030 target in 2025. Distillation capacity went from 421 crore litres in 2014 to about 1,810 crore litres. Blending rose from under 1.5% to 20%. E20 became the mandated national fuel on 1 April 2026. By any measure of industrial execution, this is among the more impressive things the Indian state has done in a decade.
It is also, at current crude prices, a net cost to the country. The government said so itself. In a Lok Sabha reply on 23 July 2026, officials conceded that ethanol is costlier than petrol at roughly $70-a-barrel crude. In the same reply, they ruled out bringing back ethanol-free petrol. In one document the government both admitted the fuel costs more and closed the exit.
This is not a contradiction. It is the honest shape of the programme, once you understand what kind of instrument E20 actually is.
Every number here is sourced from government filings, CEEW analysis, PIB data, or identified research. Where interpretation begins, I will say so.
Twenty Percent, No Opt-Out
E20 means every litre of petrol sold at every pump in India contains 20% ethanol by volume. The ethanol is blended into petrol at oil marketing company depots before the fuel reaches the retail outlet. The consumer cannot choose a different blend. Since 1 April 2026, regular petrol is E20. The only ethanol-free option is premium petrol at roughly Rs 160 per litre. Sales of premium more than doubled by July 2026, which is a revealed preference worth noting.
The ethanol comes from three broad feedstock routes. The first is sugarcane-based: C-heavy molasses (the sugar-depleted residue left after crystals are extracted, yielding about 220 to 225 litres per tonne), B-heavy molasses, and sugarcane juice. The second is grain-based: maize (380 to 460 litres per tonne) and rice routed through the Food Corporation of India (450 to 480 litres per tonne). The third, still nascent, is second-generation: crop residue and agricultural waste.
The price is not market-determined. The government sets it annually, differentiated by feedstock, to keep distilleries and farmers viable. Oil marketing companies (IndianOil, BPCL, HPCL) are obligated to procure at these prices. The landed cost at the OMC depot works out to about Rs 71.55 to 72.75 per litre after tax and transport, depending on feedstock.
Here is the arithmetic everybody gets wrong. A litre of ethanol costs about Rs 72. A litre of petrol, at $70 crude, costs about Rs 46. Ethanol is Rs 25 dearer. That gap is the number critics cite, and it is correct as far as it goes.
But ethanol carries roughly a third less energy than petrol. In a 20% blend, the energy content drops about 6.7%. On an energy-adjusted basis (rupees per unit of useful energy, which is what actually moves the vehicle), ethanol costs about Rs 110 per litre-of-petrol-equivalent. That is the real number. Most public discussions never reach it.
At What Crude Price Does the Maths Flip?
The entire economics of E20 pivots on a single variable: the price of crude oil.
Ethanol's cost is roughly fixed by administration. Petrol's cost floats with crude. Where they cross is the break-even. A simple model, assuming Rs 85 to the dollar, 159 litres per barrel, about Rs 9 per litre for refining margins and transport, and ethanol fixed at Rs 71.55, reproduces the government's own stated break-even. The numbers:
| Crude Price ($/bbl) | Petrol Cost (Rs/L) | Ethanol Cost (Rs/L) | Gap (Rs/L) | Programme Status |
|---|---|---|---|---|
| $60 | ~Rs 41 | Rs 71.55 | +Rs 30 | Net cost |
| $70 | ~Rs 46 | Rs 71.55 | +Rs 25 | Net cost |
| $90 | ~Rs 57 | Rs 71.55 | +Rs 15 | Premium shrinking |
| $120-130 | ~Rs 71-79 | Rs 71.55 | Rs 0 | Break-even zone |
| $150 | ~Rs 88 | Rs 71.55 | -Rs 16 | Hedge pays out |
Two different break-evens need to be kept separate. The exchequer break-even (procurement cost versus petrol cost) sits around $120 to $130 per barrel. This is the number the government references. The consumer break-even is higher still, because the consumer also absorbs the mileage loss. The crude price at which the motorist is actually made whole is well above $130, probably above $150.
Every claim about E20's economics is, implicitly, a forecast of crude prices. Defenders assume high or volatile crude. Critics assume cheap crude. Nobody knows where crude goes, which is precisely why the honest framing is insurance, not savings.
In August 2026, Brent sits around $83. J.P. Morgan expects it to average $86 in Q3 2026 and $78 by year-end. The EIA is more bearish, forecasting $74 for Q3 and an average of about $65 through 2027. At all of these levels, India is below break-even. The programme is paying premium and collecting nothing, which is exactly what insurance does in a quiet year.
Who the Bill Lands On
Three groups, in descending order of visibility.
Rs 88,234 Crore, Paid at the Pump
The Reporters' Collective published the number in July 2026. It is derived from physics, not sentiment. Ethanol carries about a third less energy than petrol. Across India's total motor spirit consumption, the energy deficit from E20 meant motorists burned approximately 6.57 million tonnes of additional fuel over three fiscal years (April 2023 to March 2026), at a cost of Rs 88,234 crore. FY 2025-26 alone accounted for about Rs 37,843 crore.
ARAI's dynamometer tests found 2 to 6% fuel consumption increase from E10 to E20. The government FAQ says 3 to 5%. LocalCircles, polling over 44,000 owners of pre-2023 vehicles, found 66% reporting mileage drops exceeding 10%. The gap is explained by what a dynamometer excludes: idling, air-conditioning, a decade-old oxygen sensor, stop-start traffic. A vehicle in poor maintenance was losing mileage before E20 arrived. Ethanol is the change its owner noticed.
The cost falls hardest on old two-wheelers, gig riders, and small commercial users, the people with the least ability to absorb it. A delivery rider doing 3,000 km per month pays Rs 450-plus in extra fuel monthly; a taxi driver doing 5,000 km, nearly Rs 2,400. There is no surcharge on the receipt, no subsidy office to visit. The cost is invisible and regressive.
The Sticker Price Is Rs 24,824 Crore Short
CEEW puts the true public cost of ethanol procurement in ESY 2024-25 at Rs 87,390 crore. The OMC sticker figure is Rs 62,566 crore. The Rs 24,824 crore gap is made up of fertiliser subsidies for feedstock cultivation, electricity subsidies for irrigation and distillery operations, foregone GST, interest subvention on distillery loans, and the FCI rice under-recovery. The programme is about 40% pricier for the country than the headline number.
The subsidy is structurally distributed across at least four ministries, ensuring that no individual budget line is large enough to trigger alarm. This is not conspiracy; it is how Indian fiscal federalism works. The effect is the same: the true cost is something analysts discover rather than something the government declares.
Rice Leaves the Warehouse at a Loss
The Food Corporation of India sold 6.35 million tonnes of rice to distillers in the thirteen months to June 2026 at Rs 2,250 to 2,320 per quintal. FCI's acquisition cost was Rs 3,720 to 3,889, roughly 40% higher. The value of that spread: Rs 14,597 crore, disclosed in a Rajya Sabha reply.
The government says this is not a subsidy because the rice moved at the prescribed OMSS price. The accounting is correct. The economics are unchanged: a fixed price 40% below acquisition cost is a transfer, wherever it is booked. Ashok Gulati of ICRIER calls it "the most irrational policy that the government has."
The distance between Rs 71.7 molasses ethanol and Rs 126 rice ethanol is also the distance between a programme that scales sustainably and one that has to be defended in Parliament every session. At Rs 126 per litre, rice ethanol is more expensive than petrol at any crude price below about $170 per barrel, a price Brent has never reached in history.
Ethanol Stopped Being a Byproduct
India's ethanol story began as a byproduct story. C-heavy molasses had limited commercial use and plenty of volume. Fermenting it into fuel was circular and cheap. That story ended around 2022.
Today, of roughly 1,810 crore litres of installed capacity, 858 crore litres is grain-based against 816 crore litres from molasses. In the current supply year, grain has delivered about 480 crore litres against roughly 238 crore from cane, close to two-thirds of the total. Ethanol has gone from something made out of leftovers to something made out of the harvest.
The shift happened because molasses has a ceiling. At the blending rates the government wanted, cane byproducts alone could not supply enough volume. Grain yields are simply higher: 380 to 460 litres per tonne for maize, versus 220 to 225 for C-heavy molasses. The capacity build was fast and heavily incentivised, with interest subvention of 6% per annum or 50% of the bank's lending rate on distillery loans.
Four consequences followed. The fiscal cost climbed: the programme moved up a cost curve from Rs 71.7 molasses to Rs 126 rice. The agricultural base shifted: the Economic Survey 2026 records maize output growing at 8.77% per year while pulse acreage fell. The food chain felt it: maize-to-ethanol volumes went from one million tonnes in 2022 to over six million in 2024, pushing corn prices from Rs 15,000 to Rs 25,000 per tonne and flipping India to a net maize importer for the first time in two decades. Poultry consumed 24.2 million tonnes of Indian corn in 2025. Somewhere in this sequence, a fuel policy quietly became an egg-price policy.
The water dimension compounds everything. NITI Aayog puts sugarcane ethanol at about 2,860 litres of water per litre of fuel produced. Rice ethanol is above 10,000 litres per litre. These are unpriced inputs drawn from water-stressed states.
Four Numbers in the Government's Favour
Against all of this sits a benefit ledger that is real, if diffuse.
These are genuine numbers, verified against primary PIB data. But each deserves qualification.
The forex saving is gross. It does not net out the foreign exchange spent importing feedstock. As the programme shifted to grain, India began importing maize (0.9 million tonnes) and ethanol itself (roughly 600 million litres in 2024, up 50%). Every tonne of imported feedstock leaks forex back out. The net saving is smaller than the headline, and the government does not publish it.
The farmer payment is the programme's most potent asset: real cash reaching rural India across producing states. But the sugarcane farmer's benefit flows through the mill, and cane arrears predate the programme. For FCI rice, the farmer was paid at MSP during procurement, months or years before the grain became ethanol; the ethanol programme's marginal effect on the rice farmer is close to zero. And the Rs 1.60 lakh crore farmer figure should not be confused with distillery revenue, which was Rs 1.45 lakh crore between 2014 and August 2024, concentrated in fewer than 500 units.
The climate claim is the weakest leg. IISD-GSI puts the implied carbon abatement cost at $200 to $400 per tonne of CO2, against market carbon prices well below $50. As a decarbonisation tool, E20 is expensive. The honest defence was always energy security and farm income, not emissions reduction.
A Premium, Not a Discount
Stop treating "costlier than petrol" as a failure and read it as a premium.
E20 is, in structure, an insurance policy against oil shocks. In calm years, with cheap crude and stable shipping lanes, you pay the premium: higher fuel cost, reduced mileage, fiscal support. In a crisis, whether a crude spike or a supply disruption, the domestic supply pays out, and you are partly insulated. The Rs 25 per litre gap at $70 crude is the premium. Protection above $120 is the payoff.
Two dividends that a financial hedge cannot buy make this instrument unusual. The premium also builds domestic industrial capacity, 1,810 crore litres of distillation against 421 crore before, and generates a sustained rural income transfer. Crude futures and a larger Strategic Petroleum Reserve would hedge the price risk more cheaply and more flexibly. But they would not reverse outmigration in Bihar's grain belt or clear cane arrears in Uttar Pradesh. E20 is a physical, industrial, and social hedge bundled into one instrument.
The honest counter is that bundled instruments are inflexible. CEEW notes that Indian petrol demand is projected to peak around 2032 near 5,700 crore litres before declining toward 3,700 crore litres by 2050 as electrification scales. Distillation capacity built on 2026 assumptions with subsidised debt has to earn out against a shrinking market. At 1,822 crore litres of installed capacity against roughly 1,016 crore litres of demand at 20% blending, the programme is already capacity-surplus at its ceiling. The honest expectation is that the public balance sheet ends up holding the stranded-asset risk.
Insurance works when the premium is proportional to the risk covered. India's crude import bill is about Rs 10 lakh crore annually. The ethanol programme offsets perhaps Rs 38,000 crore of that, a few percent. The true cost, Rs 87,390 crore in public expenditure plus Rs 37,843 crore in annual consumer mileage cost, is well over Rs 1 lakh crore per year. The premium-to-coverage ratio is high. Whether it is justified depends on whether you value the non-financial co-benefits enough to make up the difference. That is a legitimate debate. What is not legitimate is pretending the premium does not exist.
The Month Brent Touched $122
In early March 2026, the military confrontation between the US-Israel coalition and Iran escalated to direct strikes on Iranian oil infrastructure. The Strait of Hormuz, through which roughly 20% of global oil passes, came under effective threat. Brent crude, drifting near $70, moved to $90 in four sessions. By mid-March it crossed $100. By the last week of March it touched $122. Forecasters modelled a peak near $135.
The Petroleum Ministry estimated Delhi petrol would have reached Rs 125 per litre at the peak. Motorists actually paid Rs 94.77, because a fifth of every litre was ethanol bought at pre-agreed domestic prices, insulated from the spike. The ministry puts the cushion at nearly Rs 30 per litre at the peak.
That number deserves decomposition rather than celebration. On 27 March, the government also cut excise on petrol from Rs 13 to Rs 3 per litre and took diesel duty to zero, a revenue sacrifice the petroleum minister described as "a huge hit." Both instruments pushed in the same direction. The decomposition between ethanol cushion and excise cut has never been published. Only one of the two was free: ethanol's administered price is a pre-committed expenditure that runs regardless of crude levels. The excise cut was a new fiscal decision that directly reduced government revenue.
Still, Brent crossed the programme's exchequer break-even. Petrol's ex-refinery cost was roughly Rs 73 to 75 per litre at $122 crude, above the ethanol procurement cost. The programme was saving money, not spending it. For four to six weeks, the insurance paid out.
Then crude retreated. Below $100 within three weeks. Back near $83 by May. The programme returned to its quiet-year mode, paying premium and collecting nothing. This is exactly how insurance works. The premium feels like waste until the payout arrives. The payout feels like vindication until the next quiet stretch begins.
The government deployed the episode aggressively but made a communication error that by now should be familiar: the ministry presented the one-month payoff as the permanent condition rather than the exception, claiming the programme "saves money" without acknowledging it costs money in every month when crude is below $120. The March episode was an opportunity to build trust by saying, honestly, "We paid a premium for years and it paid off in March. That is how insurance works." Instead, the messaging implied savings are the norm. Anyone with a fuel log and a calculator can refute that.
The Cost Is Personal, the Benefit Is National
The programme moves money through three channels. The first is a broad rural flow: Rs 1.60 lakh crore to farmers, spread across millions of cane and grain growers. Genuinely diffuse, and the programme's strongest political asset. The second is a concentrated industrial flow: Rs 1.45 lakh crore to fewer than 500 distillery units, many affiliated with sugar-industry families, operating on guaranteed prices, guaranteed buyers, and subsidised capital. The third is a diffuse consumer cost: Rs 88,234 crore in mileage-related extra fuel expenditure, falling disproportionately on old-vehicle owners and gig workers. Invisible, regressive, with no receipt and no subsidy office.
The distributional picture is uncomfortable. The programme takes from the many (motorists, through mileage loss) and from the exchequer (through hidden subsidies), and delivers to two groups: broadly to farmers, and concentratedly to distillers. Which face the public sees depends on who tells the story.
Brazil's Pumps Ran Dry in 1989
Brazil launched Proalcool in 1975. The programme worked spectacularly: by 1985, over 90% of new cars ran on pure ethanol. Brazilian sugarcane ethanol has an energy return on investment above 6.5 and is among the cheapest biofuels in the world. Indian molasses-based ethanol has an EROEI of roughly 2. India operates at structurally higher cost and lower efficiency. These are consequences of geography and population density, not criticisms.
The episode India should study most carefully is 1989. International sugar prices rose. Brazilian mills shifted production from ethanol to sugar. Pumps ran dry. Consumer trust, built over a decade, collapsed in months. Sales of ethanol-only vehicles fell from over 90% to under 5% within years. The lesson: the government can mandate a fuel, but if supply falters once, the public remembers.
Brazil recovered through two innovations. Flex-fuel vehicles, introduced in 2003, eliminated lock-in by letting consumers choose at the pump. RenovaBio, launched in 2017, replaced direct subsidies with a market-based carbon credit system. India has neither. The Indian programme operates on administered prices with no carbon-intensity gradient. A distillery producing ethanol from high-cost rice gets the same treatment as one producing from efficient molasses.
Who Holds the Bag When Petrol Demand Peaks?
India has built 1,822 crore litres of distillation capacity on subsidised debt. If petrol demand peaks around 2032 and declines toward 3,700 crore litres by 2050, ethanol demand at 20% blending drops from about 1,140 crore litres at peak to 740. More than half the installed base becomes surplus.
A distillery built in 2023 on a 20-year loan expects to operate profitably until at least 2043. If petrol demand peaks ten years into that loan tenure, the distillery faces a shrinking market. The interest subvention de-risked the initial investment but does not cover the demand risk. If the lending bank is a public sector institution, the taxpayer bears the ultimate loss through recapitalisation or write-off.
Going to E30 or E85 does not solve this. It changes the timing. E85 has its own problem: ethanol at that concentration carries a 25 to 30% mileage penalty, and would need to sell at Rs 65 to 70 per litre to be energy-equivalent to petrol at Rs 102. That is below the OMC procurement cost. The subsidy required would be larger than the current E20 subsidy, and the programme's cost already exceeds Rs 1 lakh crore per year.
Betting the ethanol programme's future on EV failure is a bet against the direction of the global auto industry and Indian policy itself.
Five Fixes Already on the Table
The programme's difficulty was never the chemistry or even the economics. Both are defensible. The difficulty was communication. E20 was sold as savings when it is insurance. The mileage cost was understated. The fiscal cost was left for think tanks to discover. The choice was removed without building the case for why the cost was worth paying.
Five things follow, all of them available.
First, publish the total public cost annually, as CEEW recommends. Let the Rs 87,390 crore figure be the government's number, not something critics get to reveal. The true cost is about 40% above the procurement headline. Owning that number is cheaper than defending the gap.
Second, address the regressive incidence. NITI Aayog's own 2021 roadmap recommended a price reduction of Rs 3.5 to 5 per litre for E20 to compensate for the lower energy content. The recommendation was never implemented. The programme's costs fall hardest on a bounded, identifiable, and shrinking group: pre-2023 vehicle owners, gig workers, two-wheeler commuters. A fuel card, a maintenance subsidy, or an accelerated scrappage incentive would address the distributional core. The cost would be finite and declining as the fleet turns over. The gesture would matter as much as the money.
Third, fix the feedstock mix. Phase down the FCI rice route, the most expensive (Rs 126/L true cost), most water-intensive (above 10,000 litres of water per litre), and most fiscally opaque feedstock. Invest in maize productivity: Indian yields average 3.5 tonnes per hectare against 11 in the US. Getting to six or eight tonnes per hectare through better seed technology would make maize competitive without hidden subsidies. Create transparent allocation mechanisms so ethanol demand does not squeeze the poultry feed chain.
Fourth, invest seriously in second-generation ethanol. Crop residue, agricultural waste, and municipal solid waste are the only feedstocks that scale without touching the food chain or the water table. India burns tens of millions of tonnes of crop stubble annually in Punjab and Haryana. The technology gap is narrowing. The EU has already capped crop biofuels at 7% and is phasing out high-ILUC-risk feedstocks. India risks repeating a decade-old European mistake if it locks in first-generation capacity without a credible 2G pathway. IOC's Panipat and HPCL's Bathinda 2G plants need to reach commercial scale.
Fifth, hold at E20 and optimise rather than escalating. The programme achieved its target five years early. The domestic capacity is built. The supply chain is operational. The next step is not more ethanol; it is better ethanol: lower-cost feedstocks, higher-efficiency production, transparent accounting, equitable cost distribution. The urge to declare the next target before the current one is mature is a feature of Indian policymaking that has created problems across sectors.
Two True Things at Once
India bought a reasonable hedge against an exposure it could not afford to leave uncovered. In March 2026, the hedge visibly worked. The programme's execution, taking blending from 1.5% to 20% in a decade and building 1,810 crore litres of domestic capacity, is a genuine achievement of industrial policy.
The programme is also, at today's crude prices, a net cost to the country. The true cost is about 40% above the headline number. The mileage penalty falls hardest on those least able to absorb it. The feedstock shift from waste molasses to grain has created food-chain competition, forex leakage, and a rising cost curve. The stranded-asset risk is real and currently unpriced.
Both things are true simultaneously. The mature position is not to pick one and dismiss the other. It is to acknowledge the cost, price it honestly, address its distributional impact, and prepare for the demand curve that will, sooner or later, begin declining.
The question now is whether the government can make the honest case for the premium (who pays it, how much, and why) before the next quiet year persuades the public it was never needed at all.
Sources
- After E20: India's Ethanol Blending Programme Next Phase. CEEW (Council on Energy, Environment and Water), 3 July 2026
- In 3 years, Indian motorists paid Rs 88,234 crore extra to fuel Modi Govt's Great Ethanol Experiment. The Reporters' Collective, 29 July 2026
- Report of the Expert Committee: Roadmap for Ethanol Blending in India 2020-25. NITI Aayog (with the Ministry of Petroleum and Natural Gas), Government of India, June 2021
- Ethanol costlier than petrol when crude at $70/barrel, but saves forex & boosts energy security. ThePrint (Udit Bubna), 23 July 2026
- Rajya Sabha Unstarred Question No. 1013, 'Rice Supplied to Ethanol Manufactures', answered 28 July 2026. Rajya Sabha / Ministry of Consumer Affairs, Food & Public Distribution, Government of India (sansad.in), 28 July 2026
- Economic Survey 2025-26, Chapter 6: 'Agriculture and Food Management: Raising Productivity, Securing Incomes and Ensuring Food Security', Box VI.2, 'Ethanol Pricing and Cropping Incentives: Emerging Trade-offs for Food Security'. Ministry of Finance, Government of India (indiabudget.gov.in), January 2026
- Rice, maize, or sugarcane? India's ethanol push triggers debate over agricultural concerns, subsidies. ThePrint (Udit Bubna), 23 June 2026