GSX Monthly Coffee Report

The July 6th price explosion to $3.57 has been analyzed in depth elsewhere.

Succinctly, the market had found support based on past rains, harvest delays, and ICE NY cert stock declines. Our view is that quarterly portfolio rebalancing and new money forced short-covering by arbs and reportedly a couple of large hedge funds. Naturally, vols blew out, followed by subsequent gamma hedging, and with Brazil as the only significant origin seller, prices exploded. ICE responded by raising margins to over $21,000/contract.

For the remainder of July, NY traded sideways, mainly within a $3.10 - $3.45 range.

Icafé provided forecasts from 14 respected statistical sources, with a median forecast of 72.15 million bags.

ICE NY certified stocks declined by 100,000 bags, closing the month at 264,000 bags. The September/December NY spread firmed from 7.5 cents to more than 24 cents during July.

Certified stocks have traditionally served as the trade’s final backstop against supply delays. With destination stocks at historically low levels and freight disruptions related to the Middle East, current certified-stock levels do not represent a viable buffer.

Rains on July 24/25th in Sul de Minas and parts of the Cerrado measured 25 - 40 mm. This is perhaps sufficient to sponsor a first flowering in 1H August.

NOAA/CPC ENSO Statement (July 9, 2026) Next scheduled update 13/8:

Additionally, 2H 2026 to 1H 2027 is stacking up to have four possible headline weather anomalies: the potential super El Niño, aligning with a strong negative Pacific Decadal Oscillation (PDO), an Atlantic Niña, and a positive Indian Ocean Dipole (IOD), all sitting on record global ocean heat. We are hard-pressed to find analog years for comparison.

Excluding the Atlantic Niña, and trading off a very strong El Niño and a positive IOD with a bit of a negative-PDO filter, the risk squarely points toward robustas. The focus would be Indonesia, Vietnam, and Espírito Santo in Brazil. After prior El Niño events, robusta farmers made significant irrigation investments, especially in ES, but irrigation isn’t drought-proof.

NY Structure

Considering the 1H July ICE margin requirements, it is not surprising that NY’s futures open interest has declined. However, futures OI has been in decline now for almost two months.

NY certs dropped precipitously in July, finishing at 264k bags, of which 70% are transitional certs, not EUDR compliant. The majority of the certs changed hands against the July contract. Do the new owners plan to decertify, entering the transitional certs into the EU prior to the December 30th deadline? Or... often stoppers will buy two spreads (Jul/Dec), and if the coffee’s unsold - considering Sep/Dec exploded - perhaps shift hedges back to Sep, reissue, and pass the baton?

As mentioned before, cert stocks traditionally are the trade’s final backstop for any supply delays. Considering stocks at destination are historically low, the current cert levels aren’t a viable buffer, not even close. Besides, they’re arguably the cheapest coffee. Adding ambiguity to the NY structure is a rumor of a Brazil charter to the Sep NY board. To quote the movie Wall Street, "If you’re not inside, you’re outside." Perhaps most traders are best protected by avoiding September.

The well-capitalized Brazilian farmers played the market masterfully, not chasing the market down in late May and early June when it dropped another 40 cents. They benefited from delays in harvesting and the tight period into the new crop. Since the June 9th lows, non-commercials and index funds have bought nearly 27k lots in NY. Now, the index component’s +11k lots may be ‘sticky’, but it is not certain they’ll buy another 11k lots.

The internal Brazil farmgate arb between conilons and low-grade arabicas continues to encourage domestic roasters to use robustas. We’re transitioning from a deficit to a significant global surplus of 10 mln bags, Brazil-driven and predominantly arabica-based. How long can the Brazilian farmer wait, and only sell this huge crop into strength?

We have all seen “face-ripping” short-covering rallies before. But when the upward price move is $2.38 to $3.57 - a 50% revaluation in just under four weeks - it’s left some questioning their balance sheets. If the crop estimate is [X], why is the diff trading at [Y]?

The initial focus seems to be on conilon production estimates. Even if the crop estimates are accurate, the volatility and the ICE margins have reduced market participation. We transition into a surplus year, with Brazil the dominant supplier during 2H 2026. The global balance sheet suggests the surplus will result in an arb adjustment, and a super El Niño could well be the catalyst. Of course, one could have said that two months ago. Timing is everything.

Meanwhile, we watch the NY cert stocks erode on a daily basis. Origin is well-capitalized but will certainly sell size on rallies, and roasters hope to price and rebuild stocks, but at significantly lower prices. We have the critical flowering season ahead, and Brazil elections in October.

"As always, the only certainty is uncertainty."

Whether you're looking for weekly market updates or comprehensive commodity research and technical analysis, there's a GSX Research package to suit your needs.

Basic – Essential weekly market intelligence.

Premium – Full research reports, technical analysis, proprietary data and cross-commodity insights.

Exclusive – Bespoke research, monthly strategy calls and direct access to GSX analysts.

👉 View Plans & Subscribe Now

Arabica Forward Curve:

Robusta Forward Curve:

Flat price:

Physical prices Brazil:

CoT:

Certified stocks: