You Can’t Build a Peaker Refinery
- Crack spreads reached all-time highs as war and policy pull refining capacity off the global market
- U.S. refineries are running flat out and deferring maintenance, making one of the market’s few remaining sources of responsive supply increasingly fragile
- The clash between nationalism and capitalism means higher prices can’t summon more supply when countries keep it at home
Anyone trying to book travel this holiday season is probably asking themselves how flights could be so expensive. And not just holiday expensive. Like maybe I’ll drive expensive. And no, it’s not because AI got ahold of the dynamic pricing model. It’s that fuel prices are up over 115% from the beginning of the year and there is little relief in sight. Like we wrote last time, the world has not run out of crude. We solved that over the last 50 years. The problem is what happens next, turning that crude into diesel and jet fuel. There is a strategic reserve for oil, not for diesel.
Over the last year, the global refining system has become much less global. Partly caused by war damage and the rest caused by policy. Trying to quantify what has been lost seems like a fool’s errand. There are so many numbers out there telling you what is missing or offline, but the truth is, no one really knows how long it will take to fix damaged refineries, when large refining countries will export again and if/when flows through the Strait will return to normal. What we do know is the crack spread (spread between the cost of a barrel of crude and the value of the products that can be produced from the barrel) has blown out. It is roughly twice its five-year average right now. That did not happen during the Gulf Wars. In 1990, the crack spread peaked at roughly 31% of the crude price. Around the 2003 invasion of Iraq, it reached about 22%. In July cracks peaked ~87%.
Historically, refining has been a pretty awful, low margin business. Capital intensive. Cyclically uncertain. Environmentally liable. No pricing power. And no choice of product mix. By the late 1990s, the consultants had come for the refineries too. What had once been treated as strategic capacity became an underperforming line item. Close the small plants. Merge overlapping systems. Sell the non-core assets. Push more barrels through the facilities that remained. The strategy worked. American refining became a much better business. The system worked too, because whenever one country or region fell short, trade filled the gap.
Today, U.S. refiners are running flat out (98% utilization rate this week, which is highest in almost a decade), and the global market appears to be missing roughly 1.3 million to 1.4 million barrels of diesel shipments every day, equivalent to about 5% of global diesel demand. Much of the capacity that might normally respond is damaged, constrained or being used to protect domestic markets. Outside the U.S., nationalism has started outbidding capitalism. But not in the U.S. With margins this good, many have deferred maintenance. Which is generally fine in the short term, however, we seem to be past that. Now imagine unplanned outages remove another 2% of global diesel supply. The hole does not get 2% worse. It gets roughly 40% larger, and the competition for every available barrel intensifies.
Everywhere else in the commodity world, there is a buffer. Crude has the SPR (Strategic Petroleum Reserve). Copper and aluminum have the LME (London Metals Exchange) warehouses and the national stockpiles. Grain has the silo. Power has peaker plants. Refining has nothing. You can’t build a peaker refinery. Nobody (unless government sponsored) is going to spend $15 billion building one that runs only during emergencies. And I’m not telling you that we tore out the buffer that used to exist, because really the ‘slack’ in the system was just a glut. I’m saying every previous refining crunch was solved by trade, and that is what has changed. For years, when one region came up short, refiners somewhere else ran harder and shipped fuel into the gap. Now, with limited spare capacity and countries keeping more supply at home, shortages are becoming increasingly regional. That makes terminals, storage, racks and distribution networks more valuable. They cannot create diesel, but they can source it from more places, hold it closer to demand and move it when normal routes fail.
Trade was the reserve. Today, America is not simply running at 98% to satisfy American demand. It is filling holes left by a product market that is no longer a true global market. And if something breaks here, a higher price may not be enough to bring the missing diesel back. Capitalism will still clear the market. It may just have to do it by forcing someone to stop buying. So, when will the American refining complex break? No one knows. But with maintenance being pushed into next year, no diesel SPR behind it and fewer countries willing to export when supply gets tight, we appear determined to find out.
