On August 25, at an event in Lucknow, Uttar Pradesh Chief Minister Yogi Adityanath told an audience that there was no sugar crisis in the state. UP consumes about four lakh tonnes of sugar a month. It had about 28 lakh tonnes in stock. Divide the second by the first and you get seven. Seven months of cover. And from October 15, mills would restart and make another 90 lakh tonnes.

Reassuring and easy to report. Which is what happened.

Now look at what sugar cost that same week. On August 19, Muzaffarnagar mandi hit Rs 5,700 a quintal against Kolhapur’s Rs 5,530. The benchmark mandi of India’s biggest cane state was one of the most expensive places to buy sugar in the country, with UP wholesale up 30.99 percent in three months. So a state supposedly swimming in sugar, but where sugar happened to be among the dearest in the country.

Both these facts were reported. They were rarely put next to each other. When you do put them next to each other, you discover something more interesting than a politician being optimistic. You discover that in India, the amount of sugar in a warehouse and the price of sugar in a shop are only loosely related, and that this is not an accident.

The number moved

Three days before that speech, on August 22, the Chief Minister reportedly held a review meeting. The government said mills in the state were holding 179.38 lakh quintals of sugar. Of this, 23.60 lakh quintals sat with cooperative mills, 5.28 lakh quintals with state corporation mills and 150.50 lakh quintals with private mills.

Convert that and you get 17.94 lakh tonnes. Not the 28 he mentioned later.

Where do the extra 10 lakh tonnes come from? The likeliest answer is in the same statement, which also noted that mills had sold 235 lakh quintals of sugar in June and July, and that officials believed much of it was still floating around in the open market. 

But sugar that has already been sold is a different animal from sugar sitting in a mill godown. For two reasons.

One, UP mills sell all over India. The state is landlocked, and it sends most of its surplus to deficit regions elsewhere in the country. A good chunk of what left UP mills in June and July also left UP.

Two, sold sugar sits with traders. And traders were, that same week, under the scanner. District magistrates had been ordered to physically verify stocks at mills and godowns, and in Meerut division alone the deputy sugarcane commissioner said 2.69 lakh tonnes were under watch. The same sugar cannot be a comforting cushion and a suspected hoard at the same time. 

Wrong again?

Now grant the Chief Minister all 28 lakh tonnes for the state. The sum still does not work, because the denominator is wrong.

UP does not eat what it makes. In 2025-26, according to state government figures, 121 mills crushed 877.96 lakh tonnes of cane and produced 89.68 lakh tonnes of sugar. Against the state’s own consumption of roughly 48 lakh tonnes a year, that is close to double. Those warehouses are not holding just the state’s sugar.

Every month, the Union Food Ministry announces a release quota. It decides how much sugar all of India’s mills, put together, may sell in the domestic market that month, and then it splits that quantity mill by mill. For August 2026, the number was 22.50 lakh tonnes across 585 mills. UP’s share was about 8.43 lakh tonnes. Maharashtra got 7.75 lakh tonnes.

So a mill can be full to the roof and still be legally barred from selling more than its allotted quantity. And in August, the ministry went the other way too, warning of punitive action against mills that sold less than their quota. So the union caps how much a mill may sell, and can also direct it to sell.

From September, the leash is likely to get shorter. Reports suggest the union food ministry has switched to fortnightly quotas, told mills to sell at least 40 per cent in the first week, and to dispatch sold sugar within seven days. The first allocation, announced on August 30, was 13 lakh tonnes for September 1-15.

So the right question is not how much sugar is lying in UP. It is how much of it is allowed out this month. And that is decided in Delhi.

The chain from the field to your tea

It helps to walk backwards from a shop to a cane field, counting the points at which somebody other than the buyer and the seller decides what happens.

The shopkeeper is free. The wholesaler is not: since August 1, under Section 3 of the Essential Commodities Act read with the Sugar (Control) Order, 2025, no dealer may hold more than 4,000 quintals, or hold any consignment longer than 30 days. From September 1, bulk buyers such as biscuit and soft-drink makers cannot hold more than 15 days’ worth.

The mill cannot sell below Rs 31 a kg. That is the Minimum Selling Price, introduced in June 2018 at Rs 29 and raised to Rs 31 in February 2019. 

The mill cannot sell more than its quota, as we have seen. It cannot import cheaper sugar to undercut anyone, and it cannot export to escape a glut. Both are policy switches. On May 13, the DGFT banned sugar exports until September 30. On August 20, it opened a duty-free quota for 10 lakh tonnes of imported raw sugar until October 31.

The cane was not bought at a negotiated price either. The Centre fixes a Fair and Remunerative Price, Rs 365 a quintal for 2026-27, up from Rs 355. UP stacks a higher State Advised Price on top, Rs 400 for early varieties and Rs 390 for common ones in 2025-26. A Constitution Bench of the Supreme Court settled in April 2020 that the state may fix prices as long as it is above the Centre’s.

And the farmer does not choose the mill. States carve the cane belt into reserved areas, one mill per zone. The mill issues supply slips, or parchis, now sent to farmers’ phones. Cane with a slip goes to that mill, and the mill must buy it and pay within 14 days. 

So several permissions between a cane field and your morning tea. Only three things are genuinely free. The price above Rs 31. Which buyer a mill picks. And the retail margin.

One more thing worth noting. None of this is dusty machinery nobody bothered to repeal. The Sugar (Control) Order, 2025, notified on May 1 last year, replaced the 1966 order. The architecture was reviewed, with the floor price, the quota and the stock limits retained.

Why everyone would rather wait

Here is the number that explains the behaviour of the market better than any conspiracy theory.

The floor price of sugar has been Rs 31 a kg since February 2019. Over the same period the cane price has climbed from Rs 275 to Rs 65 a quintal. ISMA, the mills’ association, says the cost of making sugar is now above Rs 41 a kg. The cooperative federation has asked for an MSP of Rs 41. So has the Maharashtra government. Industry people quoted by Business Standard put the cost at Rs 42-43 a kg and say mills have spent much of the last two seasons selling below it.

The Food Ministry did physically verify stocks and found some mills holding more than they had declared, others selling less than their quota, and sugar sold at the start of a month being lifted only at the end of it. It concluded that the surge was driven by hoarding and speculation. That is fair as far as it goes. But hoarding here is a response to an incentive the price structure itself creates.

Mills crush from November to April and must pay farmers within a fortnight. We eat sugar all year. Without a cap, every mill would dump its output into the same twelve weeks to raise cash, and prices would crash every February.

India has tried the alternative. In April 2013, acting on the Rangarajan Committee’s advice, the Cabinet scrapped the levy obligation and the release mechanism. It lasted five years. And it was only half a reform, because the selling side was freed while the buying side was left alone. Cane prices kept rising regardless of what sugar fetched. When the glut of 2017-18 arrived, mills could not cover costs, arrears exploded, and in 2018 the floor price and the monthly quota came back.

The lesson was that freeing one end of a two-ended price control is worse than either the control or the market.

The crop that did not show up

There is one more piece to this.

In November 2025, ISMA expected net production of 309.5 lakh tonnes and a closing stock of 74.5 lakh tonnes. On that basis the government allowed exports. By February, the estimate was cut sharply. The season is now expected to end at about 279 lakh tonnes with a closing stock near 35 lakh tonnes, against consumption of 280-285 lakh tonnes. Roughly six weeks of cover. This happened in a record cane year. The Agriculture Ministry’s third advance estimates put 2025-26 sugarcane production at a record 500.06 million tonnes.

How can there be record cane and disappointing sugar? UP shows how. Cane delivered to mills there fell from 937.63 lakh tonnes to 877.96 lakh tonnes, even as recovery improved from about 9.7 per cent to 10.21 per cent. Officials blamed a warm February, unseasonal March rain, red rot in the Co-0238 variety, and diversion of cane to khandsari units that pay cash on delivery.

A farmer can take cane to a kolhu, a khandsari unit or a juice crusher instead of a mill. That road has no quota and no reserved area. Of the 500 million tonnes of cane India grew, mills crushed roughly 300 million tonnes. The rest went to seed, fodder, jaggery, khandsari and waste. The USDA’s India sugar tables lump gur, seed and feed into a residual bucket. Which means every forecast of Indian sugar output rests on mill data, while a large slice of the crop moves through a channel nobody measures properly.

The government did try to fix this. The draft Sugarcane (Control) Order, 2026 would have licensed kolhus and crushers, made cane price payment compulsory for khandsari units, and widened the distance between mills from 15 km to 25 km. Farmer unions called it anti-farmer. The government withdrew the draft at the end of May.

What a CM can and cannot do

Sugar sits in the Concurrent List. Rule-making, which means quotas, floor prices, stock limits, imports and exports, is central. Cane pricing, reserved areas and supply slips are state subjects. So is enforcement, and enforcement matters, because the Food Ministry has no field staff and a stock limit without a district magistrate behind it is a press release.

But enforcement is execution, not policy. A state can make a central rule bite. It cannot change the rule, and it cannot add a single tonne of supply or release stock faster than Delhi permits. That was the position Uttar Pradesh was actually in during the last week of August. 

India is not unusual in regulating sugar. The United States still runs marketing allotments that cap how much sugar each processor may sell for domestic use, backed by price-support loans and import quotas, and Congress requires the Agriculture Department to manage supply tightly enough that the programme costs the taxpayer nothing. The European Union ran production quotas and a guaranteed minimum beet price from 1968 until October 2017. When it finally let go, output rose, prices fell, beet acreage shrank and factories closed across France and elsewhere. Sugar is a product governments find very hard to leave alone, and those that do let go have not always enjoyed the result.

Where India differs is in which end it regulates. Brazil, the largest producer, deregulated in the 1990s, and mills and growers then built their own pricing system in the vacuum. Under CONSECANA, cane is paid for according to its sucrose content and the realised prices of sugar and ethanol, so when sugar falls the cane bill falls with it. Thailand keeps a statutory 70:30 split of sugar and molasses revenue between growers and millers; challenged by Brazil at the WTO, it scrapped its domestic price control and sales quota in 2018 but kept the sharing formula, and the dispute was settled in 2024. Its growers now complain that the split ignores ethanol and power revenue.

So where does this leave us?

Prices have started to cool. Ex-mill rates are down about a fifth from the peak, and Muzaffarnagar was quoted at Rs 5,350-5,400 a quintal on August 29. October output is expected to cross 10 lakh tonnes with crushing brought forward to October 15, and that October sugar can be sold without quota restrictions. Retail prices, which always lag, had not yet followed.

As prices soared, sugar mills earlier planned early crushing. The proposal came after sugar prices rose amid concerns that below-normal monsoon rainfall could affect cane production. 

ISMA and the cooperative federation warned that starting the season early will reduce recovery and yields, and asked for compensation. 

Amid the close of the season, ISMA has now reiterated the need for an early revision of the MSP of sugar, citing rising production costs and subdued ex-mill realisations. In Maharashtra, cane arrears stood at Rs 2,130 crore as of mid-April, significantly higher than Rs 752 crore recorded during the same period last year.

Real reform would have to move the cane price, the state advised price, the reserved areas and the release mechanism together, because we already know what happens when you free one end and legislate the other.  Uttar Pradesh votes in early 2027, and there is a limit to how far any government can push sugar prices down before then without cane dues piling up again. So the controls stay. 

So the next time a full godown is offered as reassurance, ask a different question. How much of that sugar is allowed out this fortnight, and who decides? For a shopper in Muzaffarnagar, as for one in Chennai, the answer is the same, and it is neither the person selling it nor the state it happens to be sitting in.

We spent a month arguing about the size of the tank. For the person at the counter, the only thing that ever mattered was the width of the pipe, and who holds the tap.