Jeffrey Currie (executive co-chairman, Abax Markets; senior advisor, Carlyle; ex-Goldman Sachs global head of commodities research) · oil, commodities, energy flows and positioning — running synthesis of his posts and appearances, with per-item breakdowns and a stock index.
His own pre-IPO venture (co-founder, non-exec director) — "the cheapest asset inside the cheapest thing on your screen": Gulf shallow-water reserves bought near $4/bbl, the Renaissance acquisition (11 fields, 23 platforms, 88 wells, ~3,000 boe/d, ~250 miles of pipe) as the base for a cash-first roll-up paying a 5% dividend, run by Talos builder Tim Duncan.
The overlooked shortage — Ukrainian strikes on Russian supply and Black Sea grain shipping, disappointing US yields, fertilizer problems and a super El Niño: "I'd be looking at agriculture as well."
The picks-and-shovels leg of the shallow-water thesis — Currie holds a board seat and is "a big believer": the highest-quality jack-up fleet drilling the marginal barrels in the Middle East, the Gulf of America and Asia.
The immediate trade — "as soon as I'm off this call, I'm buying gold." Short gold March→June on rate-hike fear, now long the shift: gold is the current bottleneck in his rotation and the hedge against a Hormuz-driven end of the Bretton Woods bargain. "Gold could go to 10,000. I'm in that camp."
Structurally long and not adding at the margin, but the action has left crude: cracks at $60 and $83 on one day — more than the price of crude — say the world is short refineries, not barrels. "Nobody consumes crude oil": buy the petroleum indices, or refiners / producers / integrateds.
The higher-beta leg of the precious call — "I bet silver is probably the one that's going to rip in this next spike" and only 7–8% into the move; "silver could go to 300 or something ridiculous like that."
The cheap, dividend-paying "grandma" oil major — down to ~$140 from ~$170 — whose long-term outlook "has probably never been better": under-investment, robust demand, lost Middle East supply/refining; the dividend is "the real yield."
Named only in passing as a hot tech name ("up 7%… up 10% before we hopped on") to argue the tech story is running out of steam and the space is overvalued — no investment view.
Currie's bellwether of tech over-investment — free-cash-flow yield "likely to go negative" as AI capex burns cash, the same 120%-of-cash-flow signal oil producers flashed at the 2014 top: "rising ROC will be the oil guys, not the tech guys."
In one line: "Get long, buckle your seatbelt and hang on for the ride." The oil bull market "is just getting started" — physically tight (5–6 mb/d draws, SPR at a 43-year low) yet financially abandoned, so the cheap dividend-paying majors are a buy here, and the bigger trade is the "revenge of the old economy": rotating trillions of capital out of cash-burning tech into under-invested molecules (oil/gas), atoms (metals — copper "the new oil") and the electrons they feed. By July the situation has turned "pretty dire": round two of the Hormuz/refining shock with the insurance policies exhausted — and the acute shortage is now in refined products, not crude. By August the message is sharpened to "nobody consumes crude oil" — crack spreads at $60 (and $83 on one day, more than the price of crude itself) say the world is short refineries — and the method is explicit: the bottleneck rotates (crude → products → copper → gold/silver), so you bank the spikes. Fresh money is going to precious metals and agriculture; energy he simply already owns. By September the crack call has an explicit trade and a turn: US diesel at a record $5.90/gal implies Brent ~$135 against ~$97 spot, so the dislocation is in the margin — but China is back, Shanghai futures traded over $100, and he wants to be long crude (stopping one word short of short diesel cracks) as Chinese refining restarts normalizes the spread. And on the producer group: "if you don't have spare capacity, it's no longer a functioning cartel by definition."
The crack, not the barrel — and the marginal processor closes it (2026-SEP-08). Back out the crude price the product implies: US diesel at an all-time-high $5.90/gal implies Brent ~$135 at normal refining margins against ~$97 spot, so the abnormality is the conversion margin. Origin is Russia — Ukrainian drones striking 1,300 km inland took out >50% of Russian refining at one point (Russia = 10% of globally exported diesel; bad enough that Russia imported diesel and jet fuel from Japan and India). But the under-watched driver is China: the consensus says China pushed crude down by not buying — it also dominates refining the way it dominates processing in copper, aluminium and steel, so the same behaviour pushed diesel up. The turn is already visible: Shanghai futures over $100/bbl, China moving to capture the $107 crude-to-diesel margin before it closes, and "more of a normalization in that spread… in the coming weeks" as capacity restarts. The trade: "I don't want to go as far as to say I want to be short diesel cracks, but I do want to be long crude here." On OPEC (October output unchanged): "if you don't have spare capacity, it's no longer a functioning cartel by definition" — regaining it requires reopening Hormuz / the East-West pipeline / Bab el Mandeb / Suez, and those constraints are "bigger now than three or four months ago… growing, not decreasing." Clock: 6 months minimum even if solved tomorrow, and alternative pipelines are a 2027 story.
Watch products, not crude — and bank the rotating spikes (2026-AUG-20). "I don't understand why there's this obsession with the price of oil. Does it matter? Nobody consumes it." Cracks running ~$60 and touching $83 — more than the price of crude — "tells you there's not enough refineries out there": Russian refining bombed, capacity locked behind the straits, Chinese teapots idled, and there is no SPR for products (diesel the acute short into the European winter) while crude at ~$88 "hasn't done anything." How to own it, the Goldman house rule: "buy the petroleum indices" — or on the equity side "some refiners, some producers… why not the integrated oils?" And the structural frame: the bottleneck "changes and rotates across these different markets, but the trend is the same" (crude in March/April → products → copper → gold and silver today), so commodity returns are "a sequence of spikes" and "you're banking those spikes." His parting call is the precious space — "as soon as I'm off this call, I'm buying gold" (short gold March→June, now long; "gold could go to 10,000… silver could go to 300," silver only 7–8% into its move) — plus agriculture. Energy: "I'm just long, so I'm not going to add anything at the margin."
His own money: the cheapest asset inside the cheapest sector (2026-AUG-20). Co-founder and non-exec director of 1947 Oil & Gas (pre-IPO, "a few weeks" away, ~5% dividend planned): "oil is the cheapest thing on your screen, and inside oil shallow water is the cheapest asset" — reserves at roughly $4/bbl, the Renaissance acquisition (11 fields, 23 platforms, 88 wells, ~3,000 boe/d, ~250 miles of pipe) as the base for a cash-first roll-up under Talos builder Tim Duncan, with "growth a dirty little four-letter word." The equipment leg is a board seat and "big believer" conviction in Borr Drilling — the jack-up fleet that drills the marginal shallow-water barrel. Two supporting frameworks: everything reprices (a Calgary asset underwritten at $110 still returned ~17% at $40 oil, because steel, labour and FX repriced with it — the ones who die are "those companies like Chesapeake who own the land," and "probably the same thing with the AI guys"), and the 120%-of-cash-flow top signal now pointed at tech. On Canada: "I'm a big fan of the Canadian asset base — the question is can you get out." Tariff pressure "sure will speed up that process with First Nations," and the expectation is ~1 mb/d of extra Canadian oil on water; the one-commodity-one-customer joke is dead. On US supply: gas gets short over the next several years (Ras Laffan out, ~20% of world LNG), but oil at 13.4–13.5 mb/d is done — "that ship sailed."
Hormuz is a Bretton Woods question, not an oil question (2026-AUG-20). The 1945 grand bargain — the US keeps the sea lanes open with a 400-year inherited network of ports (Malacca, Diego Garcia), the world uses the dollar — means losing the strait "is not only the end of globalization, it's the end of Bretton Woods," and with it the exorbitant privilege: "you lose it, the cost of funding in the United States is going to go up" against a 7% fiscal deficit. Mirror case: Switzerland's 30-year fixed at ~50bp because "Swiss franc is good as gold." The live transmission channel is the yen carry — "why is Bessent fighting the yen? Because all that yen comes out of the US, interest rates go up." And read official calm as management, not information: since Carter's 1977 sweater speech the rule has been "never admit to the scarcity — create the illusion of abundance," and every president since Bush senior has used the SPR that way.
Round two > round one — the buffers are gone (2026-JUL-17). "We have exhausted all the insurance policies in round one of the Straits of Hormuz" — inventory cushions spent, China already flexed, and there is no SPR for refined products. Both sides entrenched, the Houthis control the Red Sea, Yanbu (Saudi's diversion export) exposed: an identical escalation now is "far different and far more dangerous." Beware the "abundance illusion" — round one drew inventories without a shortage because the buffers existed. The acute squeeze is products: the Iran MOU "popped the pimple" (120–150M trapped barrels crushed crude), but Ukrainian drones took out ">50% of Russian refining capacity" (years to rebuild) — diesel cracks the highest since the mid-1980s, NY Harbor diesel ~$130–140/bbl. Shortages "hit growth" (Europe 2022 lost 25% of its energy-intensive industrial output).
The oil bull market is just getting started. Inventories drain 5–6 mb/d, the US SPR is at a 43-year low, and even with the Iran MOU the market is "way worse off today than three months ago" — yet oil is back near $80 and the companies have erased their entire 2026 gain. "I'm a buyer here." Prefer the equities (they fell harder than the oil) and own the cheap, dividend-paying majors — Exxon at ~$140 (from ~$170), the "grandma" stock whose long-term outlook "has probably never been better."
Revenge of the old economy — capital must rotate. A decade of under-investment (sector capex ~$400B vs the ~$750B needed) means molecules and atoms can't grow supply; capital has to rotate from the new economy to the old. The tell is capex/cash flow: tech is now spending so heavily that even Google's free-cash-flow yield is "likely to go negative" — the same 120%-of-cash-flow signal oil producers flashed at the 2014 top. "Which one has a rising ROC? The oil guys, not the tech guys." Tech has quietly become cyclical (AI compute is a refinery) yet still trades at 30–40×. By July he brands it the HALO trade — Hard Assets, Local Operations (rebuild supply chains, energy security, defense): mining CapEx is down 35% from its peak while data centers, the grid, transformers and turbines all need metal — and "oil and gas to burn in them" — so commodity prices must rise "high enough to attract capital back"; the hyperscalers "are commodity guys… in a cyclical business" that "need to be re-rated."
Own the beta, get paid to wait, "copper is the new oil." Don't chase alpha — get the under-owned energy/materials exposure most investors lack (energy back to ~3% of the index). Dividends are "the real yield": cash today, a diversification + income play if inflation/rates hit tech. Capital aversion (players "chased away" by 85→120 whiplash) is itself bullish. Next leg: "copper, copper, copper… the strategically most important commodity," riding near its ~14,000 all-time high.
Posts & appearances
One dated page per item — each has its stock table, talking points, and the saved text. Newest first.