Why does India import so much edible oil?

Trade is not balanced

Start with trade because it kills the soft version of the argument. In 2024, India imported about 16.5 million tonnes of major edible oils and exported about 0.27 million tonnes. Exports exist, but they are not the offset. The chart is an imbalance chart.

That matters because the domestic debate often starts with production targets. Production matters, but the first scale check is simpler: the market is being supplied by imported oil in volumes that domestic exports do not remotely match.

Read the latest gap, then read the slope. Imports rose from about 10 million tonnes in 2012 to about 16.5 million tonnes in 2024. The export line sits close to the floor because exports are small at this scale.

The import basket is three oils

The import number is not a foggy pile of cooking oils. In 2024, palm oil alone was about 8.6 million tonnes. Soybean oil was about 4.1 million tonnes. The sunflower-safflower-cottonseed group was about 3.7 million tonnes. Rapeseed-mustard and groundnut oil barely register as imports.

This composition is the story. Palm gives cheap, scalable frying fat for households, restaurants and packaged-food producers. Soybean and sunflower-type oils serve another part of household and industrial demand. India is not merely short of “oil” in the abstract. It is short of particular oils at market scale.

The bill is the price shock

Tonnes tell you the dependency. Dollars tell you the stress. The same import basket cost about $16 billion in 2024, and it crossed $20 billion in 2022 when global edible-oil prices were hot after the pandemic and the Russia-Ukraine shock.

That is the reason this is not only an agriculture story. A bad year in global oilseed markets, a supply shock in the Black Sea, a policy move in palm-exporting countries, or a weaker rupee can move India's edible-oil bill even before the household sees the final retail price.

Domestic output is real, but it is seed

UPAg does not show a failed domestic oilseed sector. Total oilseed production rose from about 6.6 million tonnes in 1966-67 to about 43 million tonnes in 2024-25. The 2025-26 third advance estimate is near the same level.

The caveat is the unit. UPAg production is seed. Comtrade imports are oil. One tonne of soybean seed does not become one tonne of soybean oil. Mustard, groundnut, soybean, sunflower and sesame have different oil recoveryoil recoveryOil recovery is the share of a seed's weight that becomes usable oil after crushing.It is why 1 tonne of soybean seed cannot be compared directly with 1 tonne of imported soybean oil. rates. Crushing, stocks, feed use, industrial use and processing losses sit between the farm and the edible-oil market.

So the contradiction is not a data error. India can harvest a lot of oilseeds and still import a lot of edible oil because the two numbers measure different parts of the system.

Yield is the slow lever

The yield chart is the productivity story. Oilseed yield rose from roughly 400 kg per hectare in the late 1960s to around 1,400 kg per hectare recently. That is real progress, but not a clean productivity break.

This is why yield belongs near the centre of the article. If India wants to reduce import dependence without endlessly moving land away from other crops, it needs more oilseed output from the same hectare. That is harder than announcing a target. Oilseeds are often rainfed, region-specific and exposed to monsoon risk.

Land is not infinite

Oilseed area rose from about 16 million hectares in the late 1960s to about 31 million hectares in the latest UPAg estimate. Acreage helped output grow. It also shows why an acreage-only answer runs out of road.

Every extra hectare has an opportunity cost. Oilseeds compete with rice, wheat, pulses, cotton, fodder and vegetables. More acreage can help in the right places, especially when prices support farmers. But land is not an empty spreadsheet cell.

The current season looks better, but it is not a solution yet

The UPAg progressive sowing snapshot adds a live-season check. Total oilseed area sown is about 11.0 million hectares for 2025-26 in this snapshot, compared with about 9.6 million hectares at the same point last year. That is a useful pickup.

But the same chart also shows the limit of the claim. Current sowing is still below the 12.4 million hectare target. More importantly, area sown is not final production, yield or edible-oil output. Weather, crop survival, harvesting, crushing and oil recovery still decide how much usable oil arrives later.

This is the honest way to use the latest UPAg data: it says the season has started stronger on area coverage. It does not say India has solved edible-oil imports.

The domestic crop mix is different

The crop-mix chart shows why a big oilseed harvest does not automatically solve the import basket. In 2024-25, soybean, rapeseed-mustard and groundnut were the big domestic oilseed crops. Sunflower seed was tiny. Palm oil is barely present in this field-crop picture because oil palm is a plantation system, not the main seasonal oilseed crop system captured here.

So the country grows one basket and imports another. A household can switch oils when prices change. A snack manufacturer, bakery or restaurant has less freedom because texture, shelf life, frying performance and cost all matter. The market buys functional fat, not a crop list.

The state map concentrates risk

Domestic oilseed production is geographically concentrated. Rajasthan, Madhya Pradesh, Maharashtra and Gujarat sit at the top of the 2024-25 UPAg state table. That lines up with mustard in Rajasthan, soybean across Madhya Pradesh and Maharashtra, and groundnut in Gujarat.

This matters because a national edible-oil target still passes through a few regional monsoons, price signals and procurement conditions. If one big state has a weak season, the national balance tightens quickly.

The shelf starts abroad

The partner chart makes the shelf real. Indonesia and Malaysia matter because palm oil matters. Argentina and Brazil matter because soybean oil matters. Russia and Ukraine matter because sunflower-type oil matters.

This is why edible-oil dependence is also a geopolitical and logistics exposure. A port restriction, drought, war risk, shipping disruption or export policy change abroad can become a price problem in India.

The kitchen sees the shock

The final chart moves from ports to the household. MoSPI's oils-and-fats CPI and refined-oil inflation show why import dependence matters outside trade tables. Retail prices swing sharply because global oil prices, duties, inventories, exchange rates and domestic margins all feed into what households pay.

Do not read the CPI chart as a one-for-one pass-through from imports. It is not that clean. Read it as the household-facing symptom of a system where a large share of supply is priced through global markets.

How to read these numbers, and what they cannot say

This article deliberately keeps seed, oil, value and retail price separate. UPAg APY tables measure crop-year area, production and yield for oilseeds. UN Comtrade measures calendar-year customs trade in edible-oil HS headings. MoSPI CPI measures retail inflation. Those are connected, but they are not the same unit.

The derived import basket sums HS 1511 palm oil, HS 1507 soybean oil, HS 1512 sunflower, safflower and cottonseed oil, HS 1514 rapeseed, colza and mustard oil, and HS 1508 groundnut oil. The export line uses the same HS basket. The domestic production charts use UPAg Total Oil Seeds and major oilseed crop rows.

The biggest caveat is conversion. The article does not turn oilseed tonnes into edible-oil tonnes because oil recovery differs by crop and by processing system. That gap is not a nuisance. It is part of the answer.

The latest UPAg sowing chart is a current-season area snapshot. It should be read as a sign of acreage progress, not as a harvest forecast. The 2025-26 UPAg production numbers shown in the line charts are third advance estimates, not final estimates.