Advertise
JPMorgan is warning that the full agricultural impact of the current El Niño weather pattern is still ahead, with the most pronounced effects on global crop production expected to materialize from the first quarter of 2027.
According to Tracey Allen, the bank's agricultural commodities strategist, the weather phenomenon has so far only begun to influence farming output, and its footprint will deepen as intensity rises in the coming months. The warning comes as the climate shock coincides with a set of structural pressures — historically elevated diesel prices, persistent fertilizer shortages, and ongoing geopolitical disruptions to trade flows — that together are leaving several major crop-producing regions increasingly exposed.
Despite these converging risks, carryover inventories from previous harvests have continued to absorb part of the shock, providing a temporary cushion to physical commodity markets in the near term.
JPMorgan's outlook draws on forecasts from the U.S. National Oceanic and Atmospheric Administration (NOAA), which point to El Niño reaching peak intensity around November, with a 90% probability of the event qualifying as very strong throughout the Northern Hemisphere autumn and winter. The bank expects the agricultural fallout to become considerably more visible from Q1 2027 onward.
Vulnerable regions span the globe. Southeast Asia faces mounting pressure on palm oil and rice production, while India is exposed across multiple crops including rice, cotton and oilseeds. Brazil and Argentina — among the world's largest suppliers of soybeans, corn, coffee, sugar and cocoa — are also at heightened risk.
Agricultural prices have already begun to respond. Jason Hunter, a technical strategist at JPMorgan, noted that the JPMCCI Agricultural Index broke to multiyear highs in August. According to Hunter, the previous breakout levels are likely to act as support on any meaningful pullback, suggesting that the upward momentum in agricultural commodities remains structurally intact.
Source: JPMorgan