Malaysia's Stock Build Was Not a Production Story, and the Composition Proves It
Malaysian palm oil stocks rose 7.48% in August to 2.82 million tonnes, an eight-month high, on data released 10 September. The market read it as a supply glut and sold the benchmark down 1.63% to 4,885 ringgit, a third consecutive losing session. But output rose only 1.39%, or 24,903 tonnes, while exports fell 7.50%, roughly 100,000 tonnes. The export side did about four times the damage the production side did.
That distinction is not academic. A stock build caused by more oil coming out of the ground resolves itself when yields turn. A stock build caused by buyers not showing up resolves only when buyers come back.
The numbers, arranged properly
Crude palm oil output reached 1.81 million tonnes, up from 1.79 million, the highest since December. Genuine strength, but a 1.39% monthly gain is not the surge the pre-release commentary implied. Palm kernel output rose 1.97% to 430,599 tonnes.
Exports went the other way, falling to 1.29 million tonnes from 1.39 million. Against total stocks rising roughly 190,000 tonnes month on month, incremental production contributed about 25,000 tonnes of that and lost export offtake about 100,000. The rest sits in domestic use, imports and the usual statistical residue.
Two line items ran counter to the trend and are worth noting: palm kernel oil exports rose 37.22% to 112,879 tonnes, and biodiesel exports rose 22.59% to 35,240 tonnes. Neither is large enough to change the balance, but both suggest the weakness is specific rather than general.
The composition is the tell
Here is the number almost nobody led with. Crude palm oil stocks rose 15.20%, an increase of 217,092 tonnes, to 1.64 million. Processed palm oil stocks fell 1.71%, to 1.17 million from 1.19 million.
Refined product cleared. Crude piled up.
If this were a production glut, you would expect the build to distribute across the chain as mills pushed volume into refineries and refineries into storage. Instead refined inventory drew down while unrefined inventory ballooned. That is the signature of buyers not lifting crude cargoes — which is precisely what you would expect when your largest customer has run out of tank space.
India imported 1.54 million tonnes of edible oil in August, an eleven-month high, and then jammed its own ports. Discharge is running up to ten days late, at least nine vessels with around 300,000 tonnes are queued at Kandla, and refiners are cutting October-to-December purchases because festival consumption did not arrive at the volume they bought against. Malaysian crude palm oil sitting in Malaysian tanks and Indian shore tanks with no room are the same event observed from two ends.
September is tracking worse
The forward read is the uncomfortable part. Cargo surveyors put Malaysian exports for 1 to 10 September down between 11.7% and 17.5% from the same period in August.
August's export decline was 7.50%. September is currently running at roughly double that rate. If that holds through the month, the stock build does not stabilise at 2.82 million tonnes — it extends, and it extends into a period when seasonal output is typically still strong.
That is the bear case in its clearest form, and it does not depend on any forecast. It depends on ten days of shipping data that already exist.
What could break it
Three things, in descending order of reliability.
Crude oil has gone through $100. Brent and US crude both passed that level this week on attacks on shipping, US diesel hit a record high, and the barrel was on course for an 8% weekly gain. That materially improves palm's economics as a discretionary biodiesel feedstock and rebuilds an energy floor that had been removed in late August. We treat this separately because it is the most consequential development of the week.
Indonesian domestic absorption continues regardless of export demand, at a reported 10.7 million kilolitres of palm biodiesel between January and early September. That does not clear Malaysian tanks directly, but it constrains how much oil is chasing the same export buyers.
And the Indian pause may be a timing problem rather than a demand problem. If festival offtake catches up through October and November, the deferred purchases return as restocking. Nobody yet knows which it is.
The read for buyers
For Indian refiners this is confirmation rather than news. The oil you did not buy is visible in someone else's tanks, and origin now knows it.
The tactical point is about sequencing. Malaysian crude stocks at a fifteen-percent monthly build, September exports tracking down double digits, and a customer with no storage is as weak a near-term configuration as palm has offered this year. But the same week has put crude above $100 and left the 2026/27 production downgrades untouched.
Buyers with congestion to clear should clear it and treat the resulting low prices as an entry rather than a trend. Buyers reading the stock number as a reason to stay out through Q1 are extrapolating a destination-side bottleneck that is, by its nature, temporary.
Watch three things: whether the 1 to 10 September export rate persists through the full month, whether the crude-versus-refined stock split narrows in the September data, and whether the Kandla queue clears.
Forward-looking views here are analysis, not investment advice.
GLOBOIL India 2026
A stock number tells you how much oil exists. Its composition tells you why. Reading the difference is what separates a position from a guess, and it is the work that happens at GLOBOIL India 2026, the 29th edition, from 29 September to 1 October 2026 at The Westin Mumbai Powai Lake, Mumbai. The world's leading edible oil and agri-trade conference convenes with September export data landing and Q4 shipment decisions still open.
