Market Watch · September 25, 2026

Market Watch 5: Australia's Wheat Scare Eased. Indonesia's Rainy Season Slipped.

ABARES revised Australia's winter crop up 12% to 61 million tonnes, the fourth largest on record. BMKG expects a delayed rainy season across about 61% of Indonesia's land area, with rains now due in early November across most of the country. Two risks this series has tracked since July finally produced results, and they went opposite ways.

By Christopher W. Corwin · IAMElNino.com · 12 min read
← All Posts
61 Mt
ABARES forecast for Australia's 2026-27 winter crop, the fourth largest on record and 12% above its June figure.
61%
Share of Indonesia's land area where BMKG now expects a delayed rainy-season onset. Across most of the country the rains are now due in early November. Unrelated to the figure at left.
+2.1°C
Weekly relative Niño 3.4 anomaly, week centred 16 September 2026. Niño 1+2 +3.9, Niño 3 +3.0, Niño 4 0.0.

CPC holds an El Niño Advisory, with a greater than 90% chance of a very strong event and a 75% chance that October to December exceeds every event since 1950. The ocean signal keeps strengthening and keeps leaning east.

What has changed this fortnight is on land. Two of the risks this series has tracked since July have finally produced testable results, and they went in opposite directions.

The scorecard

RiskStatusWhat moved it
Australian wheatWeakenedABARES revised the national winter crop up 12% on a wetter southern winter
Indonesian planting calendarStrengthenedBMKG forecasts a delayed onset across about 61% of the land area, with rains due in early November across most of the country, and a shorter wet season
Ivory Coast cocoaUnresolvedReal dryness and pod loss reported, but West African rainfall has no clean El Niño relationship
Indian edible oilsNot an El Niño storyCurrency, freight and biofuel policy dominate the current increase
Panama CanalContinuingNo new official restriction since our 18 September edition
Corporate disclosureNewMalaysia's regulator planning to move El Niño exposure into listed-company reporting

Australia: the checkpoint that improved, and one we missed

In Edition 1 on 17 July we ran a stat card reading −26% Australian wheat forecast vs last season (ABARES), from the June crop report: wheat at 26.7 million tonnes and a national winter crop of 54.5 million.

In Edition 2 on 1 August we declined to refresh it, and said why: "ABARES publishes its next crop report in September, so there is genuinely no new Australian data this fortnight, and saying so is more useful than dressing up the old number as news." We named the September report as the checkpoint.

The checkpoint has resolved, and it resolved upward:

2026-27 forecastJuneSeptember
National winter crop54.5 Mt61 Mt, fourth largest on record
Wheat26.7 Mt29.9 Mt
Barley14.1 Mt16.4 Mt
Canola6.19 Mt7.3 Mt

ABARES attributes the revision to much better than expected winter conditions across the southern growing regions. The national number improved by 12% because the feared dry pattern did not cover the continent uniformly: Western Australia, South Australia, Victoria and southern New South Wales did well, while northern New South Wales and Queensland stayed poor.

ABARES released the revision at the beginning of September. We flagged the report in August, but missed the result in our 18 September edition. It should have led that issue.

One thing worth saying in our own defence, and only because it is checkable: Edition 1 never pinned the June downgrade on El Niño alone. It recorded ABARES also citing reduced planted area, weak wheat margins, sharply higher fertiliser costs and soil moisture that was already low before the event was declared, and noted that ample carry-in stocks watered down the export impact. A hedged call resolving toward the hedge is not a reversal. But the timing miss is ours.

Indonesia: the risk that strengthened

BMKG expects the rainy season to be delayed across about 61% of Indonesia's land area, with the rains now expected to begin in early November across most of the country rather than in October, and a shorter wet season overall. That is an official forecast of delayed rainfall, not a projection of damage, and it is the strongest new physical signal in this edition.

A delayed start compresses the planting calendar. Rice is the exposure that matters most, because roughly 30% of Indonesian paddy land lacks irrigation and depends directly on rain arriving on schedule.

Industry estimates of consequence are circulating, and they are not equally strong. Sorted by what they actually are:

EstimateSourceWhat kind of claim
Cocoa output during August and September down as much as 15% compared with JuneAPKAI chairman Arif ZamroniObserved comparison against June, not a forecast
Sugarcane output down about 25%APTRIEstimate of this year's drought and wildfire damage
Rice possibly down 10% in 2027KTNA official Zulharman DjusmanIndustry projection

The cocoa figure is the firmest of the three because it looks backward at a comparison already made. The rice number is a projection about a crop not yet planted. We are reporting them with their names attached so readers can weight them accordingly.

A fourth figure is circulating, a 2% to 8% reduction in 2027 palm oil production, attributed only to an unnamed analyst at a global trading house. It goes in the watch list below as an unverified market estimate and nowhere else. Palm oil remains a slow fuse in any case: biological effects of a dry planting season show up months later, not in this quarter's output.

Cocoa: a real weather problem, still without an El Niño diagnosis

Farmers in major Ivory Coast growing areas report below-average rainfall, small pods dropping, and a risk that the main crop finishes early if the dryness persists into October. Cloudiness and intermittent rain are complicating bean drying and quality.

That is a genuine physical problem, and it is more than cocoa had the last time we looked at it.

It is still not an El Niño story. West African rainfall does not carry the clean, reliable ENSO relationship that the Indonesian archipelago or the Peruvian coast do. The honest formulation:

Cocoa has acquired a physical weather concern. It has not acquired a proven El Niño explanation.

We made the same point in the August cocoa check, and the passage of time has not improved the attribution. A crop problem and an ENSO diagnosis are separate things, and a market that wants a narrative will merge them.

Malaysia moves El Niño into corporate disclosure

Malaysia's Securities Commission intends to ask large listed companies, particularly palm oil businesses, to disclose their El Niño preparedness and exposure.

This is the most genuinely new item in the edition, because it is a market consequence that does not require any crop to fail first. If it proceeds, companies have to quantify water management and fire prevention spending, lenders and insurers get comparable disclosures to reassess, and investors can rank plantation operators on preparedness rather than on reassurance.

Climate risk moving from the weather page to the financial statements is a different transmission channel from anything else in this series, and it can move valuations before it moves production.

Cost stack: Indian edible oils. Indian retail edible oil prices may rise a further 7% to 8% on industry estimates. The immediate drivers are a weaker rupee, freight and insurance costs, global vegetable oil prices, import dependence and Indonesia's B50 biodiesel policy. El Niño is an additional supply risk sitting behind those, not the explanation for the current increase. Exposure to an El Niño-sensitive commodity does not make every price move an El Niño move, a point this series made in Edition 4 about diesel and is making again here.

What would confirm this, and what would refute it

Would confirm the Indonesian risk: BMKG's next seasonal update holding the November start, followed by observed planting delays in the Ministry of Agriculture's rice sowing data through November and December. The test is sowing area, not price.

Would weaken it: rainfall arriving in the second half of October across Java and Sumatra, which would compress rather than displace the planting window.

Would settle the cocoa question either way: Ivory Coast and Ghana arrivals data from October onward. A genuine shortfall would still not establish El Niño causation, and we will not claim it does.

Our next Australian checkpoint: the ABARES December crop report, which carries the first harvest-based estimates rather than forecasts. We are naming it now so it can be held against us.

What to Watch

Australia is the checkpoint that improved. Indonesia is the risk that strengthened. Cocoa has a real weather problem but still lacks a convincing El Niño attribution. The useful development is not that El Niño might affect everything, but that the early forecasts have started producing results, and the results are diverging.

Data vintage. ABARES figures are from the Australian Crop Report of September 2026, released at the beginning of that month, compared against the June 2026 report this series carried in Editions 1 and 2. Indonesian rainfall timing is BMKG's forecast as reported 22 September 2026; the cocoa, sugarcane and rice figures in that section are industry estimates attributed to the named bodies and are not government statistics. A circulating 2% to 8% estimate for 2027 palm oil output is attributed only to an unnamed analyst and is excluded from the analysis above. ENSO values are the relative weekly index for the week centred 16 September 2026 and CPC's Diagnostic Discussion of 10 September 2026. Nothing here is investment advice or a trade recommendation.

Sources

← All Posts Teleconnections Dashboard →