The Sanction Whose Formula Pointed the Ceiling Up After the Market Came Back Down

In its 18th sanctions package, adopted on 18 July 2025, the European Union scrapped the round-number math that had governed the Russian oil price capprice capA rule that lets Western firms ship and insure Russian oil only if it was sold at or below a set maximum price, meant to keep oil flowing while limiting Moscow's revenue. since 2022 and replaced it with a formula. The cap would be set at 15 percent below the trailing 22-week average price of Urals crudeUrals crudeRussia's main export grade of oil; its price is the benchmark the EU cap formula tracks., Russia's main export grade, and recalculated automatically every six months. (Twenty-two weeks is about five months of prices; the review itself runs on a six-month clock.) The whole point was to take a politician's thumb off the scale: as the market moved, the cap would move with it, staying binding without anyone having to argue about a number.

That same package set the first dynamic value at $47.60 a barrel. The first six-monthly recalculation, announced on 15 January 2026 and live from 1 February (with a 90-day wind-down for contracts signed under the old cap), cut it to $44.10. The next automatic recalculation was due 15 July 2026. It was on track to point the ceiling sharply upward, toward roughly $64, and Brussels responded not by letting the formula run but by moving to freeze it. To understand why a self-adjusting rule had to be overridden by hand on its first live test, you have to look at what the formula was averaging.

When the 15 July recalculation pointed the cap upward, was that driven by the real economics of Russian oil today, or purely by a backward-looking 22-week window mechanically carrying a spring war spike that spot prices had already reversed?

The freeze is not stopping a rise the market justifies. It is stopping a rise the market has already retracted, that only the formula's rear-view mirror still sees.

On 28 February 2026 airstrikes on Iran and a scare over the Strait of HormuzStrait of HormuzA narrow sea passage through which a large share of the world's oil ships; a war scare there in spring 2026 briefly spiked prices., the narrow chokepoint through which a large share of the world's oil ships, tightened global supply. Brent, which had sat at $72.48 on 28 February, surged to $112.57 by 27 March. Urals, which had been trading around $52, ran into the $100s as its usual discount to Brent narrowed, peaking in April. Prices held high into mid-May, when Urals was still near $100. Only as ceasefire and reopening hopes firmed did the fear premium drain out: Urals fell below $65 by mid-June and to $54.54 by 7 July, essentially its pre-war level. The supply shock had round-tripped: a full loop up and back down, leaving no lasting change in where oil trades.

The market, in other words, had already forgotten the war by early July. The formula had not. A trailing 22-week average is a long-exposure photograph: it blurs a fast event across its whole exposure. As long as the spring and May weeks stay inside the window, they keep lifting the average even after spot has returned to normal.

To see where the formula pointed, we rebuilt its input. The rule averages the roughly 22 weekly Urals prices ending just before the 15 July recalculation, then multiplies by 0.85 (a 15 percent discount). Stitching together the sourced weekly and monthly prints from TradingEconomics and CREACREAThe Centre for Research on Energy and Clean Air, an independent research group that tracks Russian oil shipments tanker by tanker. and filling the gaps, the 22-week average lands around $75. That $75 is not the cap; it is the pre-discount average. The cap is 0.85 times $75, which is about $64. Because intraday peaks ran higher than our monthly proxies, the true figure could sit a few dollars above that; either way it points well up.

So the three numbers that mattered around 15 July line up like this: the frozen legal cap at $44.10, the actual market price near $54, and the ceiling the formula wanted at about $64. The formula's ceiling sat above both the frozen cap and the market. A cap above the price it is supposed to constrain is non-bindingnon-bindingWhen a cap sits above the actual market price, so no one has to break it and it constrains nothing.: no seller has to breach it, so it limits nothing.

The answer to the central question, then, is unambiguous. The upward pull came from the mechanism, not from the economics of Russian oil today. Nothing about the barrels changed for the better for Moscow between February and July; spot Urals ended almost exactly where it began. What changed was the 22-week average, which still carried the war weeks.

A simple 22-week average weights every week equally. There is no heavier recent weighting and no heavier first-half weighting; each of the 22 weeks counts the same, and a price only stops mattering once it rolls out of the far end of the window. The war weeks sit across the window's first two-thirds, peaking near its middle in April. The most recent weeks, June and July, are already back to normal. That is why the average is falling but still elevated.

Push the recalculation six months forward, to January 2027, and the window it averages runs from roughly August 2026 to January 2027, entirely clear of the war. The freeze does not fight the formula; it simply waits for the spring hump to age out of the window on its own. When the review returns in early 2027, a formula fed only post-war prices should once again produce a cap that sits below the market and binds, without anyone touching the dial. The override is temporary by construction.

Brussels did not break the formula. It hit pause until the one distortion inside it walked out the far end of the window.

One thing should temper the whole exercise. The cap only bites on cargoes that rely on Western shipping and insurance, and most Russian crude no longer does. CREA's tanker-by-tanker tracking found that in May 2026 only about 31 percent of Russian crude moved on G7G7A group of seven major advanced economies (US, UK, Canada, France, Germany, Italy, Japan) that, with the EU, run the oil-cap regime; the EU led the 2026 move to freeze the cap.-plus tankers; the rest travelled on a 'shadow fleetshadow fleetAging tankers with opaque ownership and non-Western insurance that carry Russian oil outside the reach of G7 services, and thus outside the cap's practical grip.' of aging vessels with opaque ownership and non-Western cover, beyond the cap's practical reach. So the exact dollar figure, $44.10 or $64, directly governs a minority and shrinking share of the trade.

That does not make the number meaningless. It remains the legal trigger for secondary sanctionssecondary sanctionsPenalties aimed at third parties (non-Western banks, traders, insurers) who help breach the cap; the cap number is the legal trigger they are measured against., the benchmark against which banks, traders and insurers are judged, and a signal of Western resolve. Freezing it low keeps that baseline from drifting upward on a distortion. But it does mean the freeze is as much a legal and signaling decision as an economic one, and that the cap's grip on Moscow's revenue is looser than any single figure suggests.

We set the test in advance: if a careful reconstruction of the 22-week formula landed at or below the frozen $44.10, the thesis, that the formula pointed up on backward-looking mechanics alone, would be wrong, and the freeze would look like blocking a move the numbers did not support. It did not. The reconstruction clusters near $64, about 45 percent above the freeze and roughly a fifth above the mid-$50s market. The formula clearly pointed up, and it did so because of the spring war weeks it still averaged, not because Russian oil is worth more today.