What percentage of Russian seaborne crude oil export loading capacity will be offline at end of June 2026 due to Ukrainian strikes (vs the H2 2025 baseline)?

TL;DR

My median forecast is 16%, with a mean near 16% and an 80% interval of roughly 10% to 22%. The modal path is a fallback calculation, not an engineering outage audit: Bloomberg's June 1–28 seaborne crude flow of 4.13 mb/d maps to 17.4%, while S&P Global's calendar-June flow of 4.414 mb/d maps to 11.7% (Meduza/Bloomberg, S&P Global). The March 40% outage did not persist, because June crude exports recovered to record or post-invasion-high levels as refinery damage pushed crude out to sea (The Moscow Times/Reuters, Reuters via Logistics Maritime Professional).

Context

The March disruption was real. Reuters' March 25 stocktake said Ukrainian drone attacks had knocked out about 40% of Russia's oil export capacity, or about 2.0 mb/d, with Primorsk and Ust-Luga suspended again and Novorossiysk, a roughly 700 kb/d port, lagging after earlier attacks (The Moscow Times/Reuters). The question now turns on the June 1–28 average and the source hierarchy. I found flow reporting from Reuters and Bloomberg for late June, but not a new Reuters/Bloomberg article that directly states June-average crude-loading capacity offline due to Ukrainian operations.

By June, the operating picture was different. Bloomberg said on June 2 that Baltic and Black Sea export terminals previously disrupted by Ukrainian attacks had largely resumed normal operations, while refinery strikes left more crude available for export (The Moscow Times/Bloomberg). Reuters then reported that western-port loadings from Primorsk, Ust-Luga, and Novorossiysk were expected around 2.7 mb/d in June and could reach 2.8 mb/d, above May's 2.5 mb/d, because refinery outages redirected crude to export channels (Reuters via Logistics Maritime Professional).

Evidence

The historical backbone is that actual seaborne crude exports normally run well below the 5.0 mb/d denominator. That makes the fallback formula mechanically high. In the IEA/Kpler/Argus series published May 13, the 2022–2025 annual average for Russian seaborne crude was 3.25–3.52 mb/d, so a normal year would imply roughly 30% under the fallback even without a physical port outage (IEA Oil Market Report mirror, May 13).

The formula I use for the fallback path is:

$$R = 100 \times \max(0, 1 - E_J/5.0)$$

where $R$ is the resolved percentage and $E_J$ is the June 2026 average Russian seaborne crude export flow in mb/d. Bloomberg's exact-window 4.13 mb/d gives 17.4%, while S&P Global's 4.414 mb/d gives 11.7% (Meduza/Bloomberg, S&P Global).

The repair evidence pushes against a June result near the March peak. The IEA said April Russian crude exports rose 250 kb/d month on month to 4.88 mb/d, with seaborne crude at 3.76 mb/d, after temporary late-March and early-April port disruptions; it also said Baltic loadings briefly fell from 1.5 mb/d to 760 kb/d in the week of March 30 because of a two-week Ust-Luga outage and reduced Primorsk shipments, then recovered later in April (IEA Oil Market Report mirror, May 13). CREA's May bulletin said Russian crude export volumes rose 8% month on month and Ust-Luga crude loadings rose 49% after reduced exports following repeated Ukrainian strikes (CREA May analysis, June 11). Reuters reported that May exports via Primorsk, Ust-Luga, and Novorossiysk rose to 2.5 mb/d from 2.2 mb/d in April, the largest western-port volume since September 2025 (Reuters via MarketScreener, June 2).

The June evidence is even stronger against a large continuing crude-loading outage. Reuters reported on June 17 that first-half June loadings from the three western crude ports averaged about 2.3 mb/d, about 35% above the preliminary full-month plan of 1.7 mb/d, because refinery attacks freed more crude to ship (Reuters via Baird Maritime, June 17). Reuters then reported on June 24 that western-port June loadings were expected around 2.7 mb/d and could reach 2.8 mb/d, about 1.0 mb/d above the initial plan (Reuters via Logistics Maritime Professional, June 24). S&P Global's July 2 cargo-tracking article said refinery attacks cut product exports and freed up more crude for international delivery, with June Russian seaborne crude exports at 4.414 mb/d and product exports at 1.611 mb/d (S&P Global, July 2).

My distribution puts 75% weight on a fallback-style flow resolution, split between a Bloomberg-like 17% result and a higher-flow CREA/Kpler/S&P-like low-to-mid-teens result. I put 12% weight on a late Reuters/Bloomberg engineering-capacity stocktake that resolves lower, around 5–12%, because record June loadings imply the crude terminals were mostly usable. I put 13% weight on higher source and attribution tails, where a lower June export number, residual terminal damage, short precautionary closures, tanker attacks, or denominator ambiguity push the result into the 20s or low 30s. The resulting distribution has about 11% probability below 10%, 82% from 10% to 25%, and 7% above 25%.

What's non-obvious

The strike campaign hurt Russian refining more than crude-loading capacity. That can raise crude exports. S&P Global said continued attacks on refining infrastructure cut product exports in June and freed more crude for international deliveries, and Bloomberg said the same mechanism was already visible at the end of May (S&P Global, The Moscow Times/Bloomberg). That is why gasoline shortages and refinery fires do not translate one-for-one into offline seaborne crude loading capacity.

The fallback rule can overstate true physical port damage. A record June 1–28 export rate of 4.13 mb/d still gives 17.4% because the denominator is 5.0 mb/d, not normal utilization (Meduza/Bloomberg). The IEA history shows the same point: 2024 and 2025 seaborne crude exports averaged 3.47 mb/d, which would mechanically imply 30.6% under the fallback even without a Ukraine-caused June outage (IEA Oil Market Report mirror, May 13).

Limitations

The key missing dataset is daily terminal-level engineering capacity for Primorsk, Ust-Luga, Novorossiysk, Kozmino, Murmansk, and smaller crude outlets. Public reporting gives tanker flows, selected strikes, product/refinery outages, and broad repair status, but not a daily table of berths, tanks, pumps, and pipeline feeds taken offline by Ukrainian action.

Source choice is now the largest uncertainty. Bloomberg's exact June 1–28 flow implies 17.4%, while S&P Global's calendar-June cargo-tracking flow implies 11.7% (Meduza/Bloomberg, S&P Global). A June CREA monthly bulletin or Kpler figure could land between them, and a late Reuters/Bloomberg capacity stocktake would supersede the flow proxy if it explicitly quantifies June-average offline loading capacity. That stocktake would likely pull the result lower than the fallback, but it could pull higher if it counts intermittent precautionary closures and residual damaged storage as unavailable capacity.