A veteran Wall Street strategist, known for accurately predicting the 'commodities supercycle' during the global pandemic, recently stated that the world is in the early stages of a new super-demand cycle for commodities. He suggested this cycle could persist for another decade or more, driven by the collision between the frenzy of AI computing infrastructure development and long-term underinvestment in the production capacity of raw materials like energy and metals. This view aligns with that of another major bank's strategist, who argues that the US-China strategic competition, persistent geopolitical tensions in the Middle East, and the global AI race are drawing heightened focus to core supply chain issues related to traditional energy.
The strategist, a former head of commodities research at a leading investment bank, noted on a recent program that the global energy sector presents 'the biggest asymmetric trade in modern financial markets.' This is because oil companies are offering free cash flow yields as high as 15.5%, whereas the hyperscale cloud computing providers generally lack such robust cash generation. He believes that even with supply shocks from Middle Eastern conflicts now spreading globally, this supercycle could last another 10 to 12 years.
Drawing on recent history, he observed that tech stocks dominated from the 1990s to the early 2000s, followed by energy until around 2014, before tech regained leadership. Now, however, the tech sector's insatiable demand for power and raw materials is a key driver. His latest analysis indicates that the gap between oil demand and supply is depleting inventories significantly during the seasonally weakest period for demand. This suggests that despite the recent substantial rise in crude prices, systemic pressure has not yet fully materialized.
He warned that once inventories are exhausted, prices will be forced higher to curb demand down to available supply levels, potentially replicating the historic supply-demand dynamics of 2020-2021. While benchmark crude prices have surged this year, he argued that greater pain lies ahead. The spread between spot prices and longer-dated futures indicates a fundamental mispricing of the long-term cost structure in the oil market.
He predicted that specific refined petroleum products could face critical shortages within weeks as stocks run low, leading to a 'non-linear' surge in commodity prices. Global aviation fuel is already at critical levels, Europe faces severe issues with diesel and liquefied natural gas by month-end, and the US will encounter significant gasoline supply constraints by July. He highlighted that US oil inventories are particularly tight. Concurrently, a shortage of sulfuric acid, a chemical derived from oil refining that is crucial for copper production, has helped push copper prices to record highs. Approximately 90% of the world's sulfur, the primary raw material for sulfuric acid, is recovered from oil refining and natural gas processing.
'Every policymaker, macro forecaster, senior central banker, and tech enabler is telling you everything is fine,' he remarked. 'Every commodities CEO, everyone who actually touches the physical market, is telling you that you have a real problem.'
He suggested that buying into a group of major oil companies—a playful counterpart to the 'Magnificent Seven' tech stocks—is one way to gain exposure to this record energy demand. From a pure economics standpoint, he stated, the trade is to hold these large oil firms.
The AI Era Requires More Than Chips: A Rush for Energy and Metals As resource-exporting nations like Indonesia consider new agencies to tighten commodity exports, and industrial metal prices for copper, aluminum, and nickel continue to rise this year—primarily due to surging demand for key metals and minerals essential for building AI data centers—commodities are becoming a core 'AI investment trend' beyond the immediate computing infrastructure. This trend involves investing not in models, chips, or high-performance AI servers, but in the underlying energy, metals, chemicals, and resource security premiums needed to support the expansion of AI computing capacity.
This aligns with the core thesis that the AI data center construction boom is colliding with long-term underinvestment in energy and materials capacity, potentially placing the world in a commodities supercycle lasting a decade or more.
A recent report from a major global bank suggests that as the AI construction wave transmits to commodity prices, developing nations like Chile, Peru, Brazil, Indonesia, and China stand to benefit. Chile and Peru are major copper exporters; copper is vital for AI-related electrification trends due to its extensive use in power grids, transmission infrastructure, and data center wiring. Chile also controls significant lithium resources, crucial for the dominant global energy storage technology pathway. Indonesia is the largest producer of nickel, a key input for batteries and storage systems needed to support growing power demands from AI. Meanwhile, China controls rare earth magnets critical for semiconductor manufacturing equipment, data center infrastructure, robotics, and other fields.
From an engineering perspective, AI data centers are not abstract 'clouds' but highly physical capital expenditure systems. They require a near-endless supply of efficient power for GPUs/ASICs, vast amounts of semiconductor materials and chemicals for HBM/SSD capacity expansion, and copper, aluminum, steel, transformers, energy storage, cooling systems, and natural gas/power grid support for their physical construction.
The sulfuric acid shortage pushing copper prices higher, the constraints on aviation fuel/diesel/gasoline inventories, and the undervaluation of the long-term oil cost curve, as highlighted, are essentially price expressions of the 'physical bottlenecks of the AI era.' If combined with potential policies from nations like Indonesia to centrally manage exports of key commodities like coal, palm oil, and ferroalloys to strengthen resource control, this could further amplify resource-national policy risks and supply chain security premiums.
A team led by another prominent bank strategist, noted for accurate calls, recently published a report stating that investors will continue pouring into commodities in the coming years. Even if a new Middle East conflict were to conclude, the global commodities rally could persist for years, lasting until the end of the 2030s. In their view, commodities represent the most logically clear, high-conviction 'post-conflict trade' theme. They are betting commodities will replace stocks as the biggest winners in the coming years. The core reasons are investors' urgent need to hedge against risk, inflation, and a weaker US dollar, while geopolitics and the global AI race are inherently intensifying competition for energy, rare earths, minerals, and key commodity resources.
The strategist summarized the core logic: whoever controls chips, rare earths, minerals, and efficient energy wins this global AI war. This implies that, in this bank's view, the pricing core of the post-conflict world is no longer just interest rates and corporate earnings, but rather the security of the commodity supply system, control over supply chains, and fiscal expenditure expansion.
Comments