India's sugar and E20: ethanol vs food security

In February 2026, the Indian government was still looking for mills willing to export an additional 500,000 tonnes of sugar to manage a surplus. Three months later it banned exports. At the end of July it imposed stock limits on dealers. In August, as wholesale prices hit records in parts of the country, New Delhi began considering two options that would have sounded absurd only months earlier: reducing the amount of cane used for ethanol and allowing sugar imports. The chronology looks like an energy-policy accident. It tells a subtler story: India has turned sugar into a balancing variable between its fuel tank, its food system and its monsoon.

The most tempting explanation needs to be discarded first.

E20 did not, by itself, create India’s sugar squeeze.

The crop disappointed relative to expectations, cane yields came in below earlier forecasts, end-season inventories tightened, festival demand is approaching and the 2026 monsoon revived concerns about the next harvest.

Ethanol nevertheless adds a new characteristic to the market.

Part of the cane crop can now serve two competing markets: food sugar or fuel. When harvests are abundant, that flexibility absorbs surpluses and supports mill economics. When the safety margin disappears, the same flexibility becomes an immediate policy lever for putting more sugar back onto the market.

That is exactly what is happening now.

Six months to turn the market around

On 13 February 2026, India’s food ministry was still seeking interest for an additional 500,000 tonnes of sugar exports. At that point, only 197,000 tonnes had been exported since the beginning of the season and the government explicitly presented additional exports as a way to manage surplus availability.

On 13 May, the logic changed completely.

The Directorate General of Foreign Trade moved raw, white and refined sugar from “restricted” to “prohibited” until 30 September 2026, subject to limited exemptions. Shipments already physically in the export pipeline could still proceed.

On 28 July, the government then imposed stockholding limits on sugar dealers from 1 August to 30 November, saying it wanted to discourage hoarding and speculation. Interestingly, the same official release said the recent increase in ex-mill prices was not supported by prevailing national supply-demand fundamentals.

Then came the two August developments.

On 10 August, Reuters reported that the government was considering reducing the use of sugarcane juice and B-heavy molasses for ethanol in the next season in order to prioritise sugar production.

On 18 August, Reuters reported that New Delhi was also considering limited sugar imports, potentially duty-free, possibly up to 1 million tonnes for mills, alongside the release of some stocks held by port-based refiners into the domestic market.

Those two August measures remain options under discussion, not announced policy decisions.

SOURCES: PIB/DFPD, DGFT Notification 16/2026-27, Reuters 10 and 18 August 2026.

The reversal is striking: in February the government was still seeking more exports. In August it is examining ways to put more sugar back into the domestic market.

The calculation that stopped working

The core of the investigation is less about E20 than about the lost margin for error.

In December 2025, India’s food secretary said net sugar production for the 2025-26 season was expected at around 30.9 million tonnes, after 3.4 million tonnes were diverted to ethanol. Domestic consumption was then estimated at roughly 29 million tonnes.

The cushion looked comfortable: close to 1.9 million tonnes of surplus, even before opening stocks.

That outlook could support two policies at the same time: exporting sugar and continuing to feed distilleries.

Six months later, market estimates reported by Reuters described a much tighter picture. Expected production had fallen to around 27.9 million tonnes, while consumption was put at about 28.5 million tonnes.

The expected surplus had turned into a deficit of roughly 600,000 tonnes.

These are not final audited supply-demand balances. They were produced at different points in the season and are not based on perfectly identical sources. Their value lies in the direction of travel: the market lost around 2.5 million tonnes of expected margin.

That is almost the same order of magnitude as the sugar currently absorbed by ethanol.

Ethanol was designed as a surplus absorber

Ethanol’s role in India’s sugar sector has never been hidden.

The government has explained it for years: when too much sugar is produced, part of the feedstock can be redirected into ethanol. That prevents inventories from building, supports mill economics and helps mills pay sugarcane farmers.

The Department of Food and Public Distribution publishes official sugar-equivalent diversion figures:

4.3 Mt in 2022-23;
2.4 Mt in 2023-24;
3.4 Mt in 2024-25.

The final official number for 2025-26 has not yet been published.

Available estimates do not perfectly agree. Reuters cites about 3.0 Mt of sugar equivalent diverted this season. In mid-August, the head of the National Federation of Cooperative Sugar Factories referred to about 2.4 Mt.

That difference matters. It is a reminder that a live-season estimate should not be turned into a definitive accounting number.

India’s Parliament highlighted another measurement problem on 11 August: state governments do not provide a clean physical series for the amount of cane “diverted to ethanol” because the same cane can first produce sugar and then molasses used in a distillery. The more useful concept is often sugar equivalent forgone, not simply tonnes of cane.

SOURCES: DFPD, PIB 31 July 2026, Reuters, NFCSF. 2025-26 remains an estimate.

The E20 programme has diversified. Cane remains important, but rising maize and rice use gives New Delhi room to preserve the 20% blend while returning more sucrose to the food market.

One tonne of cane can take several paths

Cane-based ethanol is not a binary process.

A mill can ferment sugarcane juice or syrup directly. In that case, a large amount of sucrose never becomes crystallised sugar.

It can produce sugar and divert B-heavy molasses, which still contains substantial fermentable sugar.

Or it can extract more sugar first and rely mainly on C-heavy molasses, a residue with less sucrose.

The hierarchy is straightforward:

juice / syrup -> more ethanol potential, less sugar

B-heavy -> compromise

C-heavy -> more sugar first, ethanol afterwards

The options reported in early August follow exactly this logic. Government could restrict juice and B-heavy use while keeping C-heavy as the main cane-based ethanol feedstock.

Official procurement prices show why switching is possible.

For ethanol supply year 2025-26, official prices are ₹65.61/litre for cane juice or syrup, ₹60.73 for B-heavy molasses and ₹57.97 for C-heavy molasses. Maize receives ₹71.86/litre and FCI rice ₹60.32.

India has therefore already built a system in which distillers can move across several feedstocks.

That makes policy more flexible than a simple “sugar versus petrol” story.

The number that prevents E20 from taking all the blame

India reached 20% ethanol in petrol in 2025-26, up from less than 1.5% in 2013-14.

The scale-up is enormous: ethanol production capacity rose from 4.21 billion litres in 2014 to around 20 billion litres in 2026. Government projects more than 12 billion litres of ethanol procurement during 2025-26.

But the composition of that ethanol has changed at the same time.

According to the petroleum ministry, maize accounts for 35.96% of ethanol production in 2025-26 and rice supplied from Food Corporation of India stocks for 24.64%. Those two sources alone account for more than 60%.

That does not make cane unimportant to the sugar economy. It means that reducing sugar-based ethanol does not mechanically break E20.

The cost moves elsewhere.

More maize and rice in distilleries means more potential competition inside those agricultural markets, even if the government says FCI rice is released only after food-security obligations have been met.

The question is therefore no longer just “sugar or petrol”.

It is increasingly:

which agricultural commodity absorbs the marginal cost of energy sovereignty?

Weather turns optimisation into a political trade-off

Ethanol works perfectly as a pressure valve while sugar is abundant.

Weather is now reducing that margin.

On 22 July, India’s Ministry of Earth Sciences said cumulative southwest monsoon rainfall was 21% below normal as of 14 July, with several meteorological subdivisions in deficit.

Government has also activated contingency planning in areas exposed to below-normal rainfall, including Maharashtra, one of the main cane-producing states.

Globally, El Niño is already active. In its 9 July discussion, NOAA’s Climate Prediction Center put the probability of a very strong event during October-December 2026 at 81%. Reuters reports that the probability was raised above 90% in the August update.

El Niño is not deterministic.

A strong episode does not guarantee drought everywhere in India or a mechanical fall in sugar output. It changes the distribution of rainfall and temperature risks, and therefore the probability of lower future yields.

For the market, that is enough.

Prices respond not only to sugar available today, but to what inventory will remain when the new crop starts and how much cane mills will have available in a few months.

The inventory paradox

On 28 July, India’s government said the recent increase in ex-mill sugar prices was not supported by prevailing supply-demand fundamentals.

It nevertheless imposed stock limits on dealers and ordered a physical verification of stocks held by sugar mills.

Those positions are not necessarily inconsistent.

A country can have enough sugar in aggregate and still experience local or time-specific tightness. Stocks may be unevenly distributed, monthly mill sales are administered, dealers may anticipate further increases, and demand rises sharply from August to November during the festival season.

On 18 August, Reuters reported that wholesale sugar in Kolhapur, a major Maharashtra trading hub, had risen roughly 20% since the beginning of the month to ₹5,350 per 100 kg.

Government may therefore be right about aggregate supply while the market is right about immediate scarcity.

The real uncertainty is the carryover stock between the old season and the new one.

That is where a few million tonnes diverted to ethanol suddenly become politically visible.

India has already almost disappeared from the world market

India’s squeeze would be a domestic story if the country were a small producer.

It is not.

The Indian government itself publishes the dramatic collapse in exports:

11.0 Mt in 2021-22;
6.3 Mt in 2022-23;
0.1 Mt in 2023-24;
0.9 Mt in 2024-25.

A country that shipped eleven million tonnes onto the world market only four seasons ago has become a marginal supplier.

The 2025-26 season was supposed to mark a partial return. Instead, it ends with an export ban.

SOURCE: Department of Food and Public Distribution, Government of India.

India used to be a swing supplier capable of changing the global balance quickly. Its withdrawal leaves the market more dependent on Brazil and Thailand.

That absence is starting to show up in prices.

On 7 August, ICE raw sugar reached 15.84 cents per pound, a four-month high, with concerns around India, Asia and European production supporting the market.

That level remains far below the peaks seen in 2023. The market is not signalling a global shortage comparable with the tightest recent episodes.

It is signalling a thinner safety margin.

Brazil can absorb the shock, but it is also choosing ethanol

Brazil remains the main global balancing variable.

Conab forecasts a 709.1 million tonne cane crop for 2026-27, up 5.3%.

Yet sugar production is expected to decline slightly, by 0.5%, to 43.95 million tonnes.

Why?

Because cane-based ethanol production is simultaneously forecast to rise 7.1% to 29.26 billion litres.

The world’s largest sugar exporter is therefore making the same physical choice as India: a tonne of cane can become more sugar or more fuel depending on prices, industrial capacity and energy policy.

The difference is scale.

Brazil currently has a much larger production cushion, and corn ethanol is expanding as well, giving it more flexibility.

Tighter Indian sugar does not automatically imply a global price spike. It means the world market has to ask Brazil and Thailand for more of the sugar India can no longer sell.

Three paths for the next season

The most important decision has not yet been taken.

It concerns how much sucrose New Delhi will allow to flow into distilleries when the next sugar season begins.

Three paths are plausible.

Scenario one: maximum priority for sugar. Government sharply restricts juice and B-heavy molasses, keeps mainly C-heavy molasses for cane-based ethanol, expands maize and rice use to preserve E20 and rebuilds food inventories. This would relieve sugar but transfer part of the energy demand into grain markets.

Scenario two: compromise. Cane keeps supplying ethanol but under a ceiling and monthly monitoring. Government uses domestic sales quotas, inventory rules and perhaps limited imports to bridge the festival season and the start of the new crop.

Scenario three: a weaker crop. Poor late-monsoon conditions or disappointing yields force New Delhi to go further: stronger cane restrictions, prolonged export controls and meaningful imports. This would be the most supportive scenario for global sugar prices.

None of those outcomes is locked in.

The lesson goes beyond sugar

India’s story looks like an agricultural story at first.

It is really about what happens when a government gives two strategic functions to the same hectare.

Cane has to pay the farmer.

It has to supply consumers with sugar.

It also has to replace some imported oil.

When harvests are abundant, those goals are compatible and ethanol acts as an excellent economic pressure valve.

When production disappoints, the hierarchy reappears.

Government must decide which use takes priority and which market absorbs the adjustment.

India appears to have already answered the first part.

It is not abandoning E20. It is trying to change what feeds E20.

That may be the most important conclusion of this investigation: energy policy is not disappearing in the face of food inflation. It is becoming more complex, shifting the marginal burden away from sugar and toward maize, rice and other feedstocks.

Risk moves as much as it falls.

For sugar markets, that means analysts now need to watch far more than cane fields.

They need to watch oil-company tenders, rice inventories, maize prices, the monsoon, export quotas and fuel policy.

An agricultural commodity has become part of India’s energy system.

And when India has to choose between the bowl and the fuel tank, the world sugar price eventually listens.

What this investigation establishes

India reached E20 while diverting several million tonnes of sugar equivalent into ethanol. That diversion was explicitly designed as a surplus-management tool. The deterioration in the expected 2026 sugar balance drove a rapid reversal in trade policy and has led government to consider reducing cane-based ethanol.

The E20 programme nevertheless has other major feedstocks. Maize and FCI rice already account for more than 60% of 2025-26 ethanol production according to the petroleum ministry. Cane use can therefore be reduced at the margin without immediately cutting the blending rate.

What it does not establish

Available evidence does not support assigning the sugar price increase to a single cause.

Weather, yields, actual inventory levels, monthly sales quotas, dealer expectations, exports already completed and festival demand all interact.

The exact amount of sugar equivalent diverted to ethanol in 2025-26 has not yet been published in a final official balance. Available estimates cluster around 2.4 to 3.0 Mt.

The sugar imports and additional cane-ethanol restrictions discussed in August remain, as of 19 August 2026, measures under consideration.

Primary sources and reference documents

Market sources used for measures not yet officially announced

Data and sources cut off on 19 August 2026.