Summary
- Industrial Info Resources’ nearly 12% outage measurement, months-long Saudi repair estimate, and Rystad Energy’s capacity assessment indicate that restart timing and substitutable idle capacity determine the speed of fuel recovery.
- IIR attributed the measured outages to attack damage in Russia, Saudi Arabia, Kuwait and the UAE plus planned maintenance, while the supplied report leaves the relative causal shares and aggregate restart date undisclosed.
- Energy analysts said Russia’s remaining refineries were already running at full capacity, making the Oct. 9 diesel agreement a likely transfer from Russia’s domestic market rather than a net addition to global supply.
- The Trump administration’s export requests, sanctions agreement and possible Defense Production Act spending change allocation and expectations without removing the offline-capacity stock on a near-term timetable.
Nearly 12% of the world’s capacity for producing gasoline, diesel and other fuels was offline as of Oct. 23, according to Industrial Info Resources, or IIR, covering attack damage in Russia, Saudi Arabia, Kuwait and the United Arab Emirates as well as planned refinery maintenance. The Wall Street Journal’s Oct. 26 report therefore described a point-in-time disruption rather than a trend. The outage volume was material, but the aggregate percentage could not by itself establish either the relative causes of the outages or their duration. The more consequential constraint was the ability to restart damaged plants and draw on idle capacity elsewhere. IIR expected affected Saudi refineries to require months of repair, while Rystad Energy’s assessment left possible Chinese spare capacity unquantified. Energy analysts said the Oct. 9 agreement with President Vladimir Putin would probably have little effect on global supply because Russia’s remaining refineries were already operating at full capacity. U.S. policy measures could shift trade and expectations faster than they could repair refineries or add domestic capacity.
The 12% figure is a snapshot, not a recovery forecast
The figure measures offline refining capacity for gasoline, diesel and other fuels. It does not inventory damage across the wider energy systems involved in the Russia-Ukraine and Middle Eastern conflicts. IIR attributed outages to drone and missile damage in Russia, Saudi Arabia, Kuwait and the United Arab Emirates, together with scheduled refinery-maintenance outages. The available report does not describe IIR’s aggregation method, disclose how the global 12% was allocated among countries, identify the geographic reach of the maintenance outages or establish whether Ukrainian refining was included in the maintenance component.
That missing detail matters because scheduled maintenance and attack damage operate on different recovery schedules. Maintenance capacity is expected to return according to an operating calendar. Damaged refineries require physical repairs, logistics and restart work. Ben Cahill, a nonresident senior fellow at the Atlantic Council, said, “Pipelines can be repaired and restored pretty quickly, but refineries and petrochemical facilities are just big, complex places and they can’t be fixed right away.” The aggregate outage figure establishes the size of the disruption at one moment, but not how much of that capacity had near-term restart dates or how much required prolonged repair.
The theaters also use incompatible denominators. IIR reported Russia’s loss as a percentage of national fuel-making capacity, while the Middle Eastern loss was reported in absolute barrels per day. Without a common allocation or capacity denominator, the Russian and Middle Eastern losses cannot be directly ranked. The nearly 12% measurement also covers several fuels, whereas Rystad Energy’s estimate of a diesel shortfall of about 2 million barrels per day applies only to diesel. The two numbers cannot serve as substitutes for one another.
Physical recovery depends on repair and substitution
IIR reported that Ukrainian forces launched more than 100 attacks on nearly 30 Russian oil refineries in recent months. The strikes knocked 19 refineries completely offline and damaged several others. IIR said the affected facilities normally had an operating capacity of 6.1 million barrels per day, about 80% of Russian output, but attacks reduced that figure to 1.6 million barrels per day. IIR estimated that the strikes removed 59% of Russia’s fuel-making capacity. Combining that reported percentage with the capacity reduction implies total Russian refining capacity of roughly 7.6 million barrels per day; the total is a derivation from IIR’s figures, not a separate IIR estimate.
The U.S.-Iran conflict produced more than a dozen direct attacks on regional refineries, according to IIR. The affected facilities had a combined operating capacity of 7.6 million barrels per day, and IIR said damage reduced fuel-production capacity by more than 1 million barrels per day. Saudi Arabia’s Jazan facility, which Houthi militants hit in September, remained completely offline. Damage to pipelines and related infrastructure slowed production and reduced daily capacity by as much as 75% at some facilities.
IIR measured a roughly 25% decline in Saudi Arabian refining capacity from attacks and logistical bottlenecks and said the affected refineries were expected to take months to repair. The report provided no plant-level restart dates, so the wider offline stock could not be assigned a recovery date. Repair complexity, rather than damage severity alone, separates facilities that can return quickly from those that require extended work.
The recovery system contains several distinct stocks. Installed refining capacity differs from available operating capacity, while offline, destroyed and damaged capacity determine how much fuel can be produced now. Russian operating capacity at the affected facilities, global diesel availability, Russian domestic fuel availability and idle capacity outside the damaged region determine how much pressure outages place on supply. Tanker movement through the Strait of Hormuz and U.S. fuel-price and inflation pressure transmit the disruption through logistics and political systems.
The relevant flows include refined-product output and exports, repair throughput, Chinese and other refinery run-rate increases, escorted tanker traffic through the Persian Gulf, and diesel-price movements feeding U.S. fuel costs and inflation pressure. Repair and logistics capacity provide the auxiliary capacity needed to restore damaged plants. Rystad Energy’s estimate of the global diesel shortfall, spare capacity outside the damaged region, and the capital requirement and production lead time for new refining capacity further determine whether other suppliers can offset the loss. The attack rate, sanctions and export-policy settings, the planned-maintenance calendar and the U.S. midterm calendar are external inputs that can change those flows.
The Russian agreement is more likely to reallocate supply than add it
The sequence matters for interpreting the agreement. Moscow suspended Russian exports in July to protect domestic supplies. Over the following months, Ukrainian strikes removed 59% of Russian fuel-making capacity. Jazan was struck in September. Separately, the Middle Eastern conflict halted tanker traffic through the Strait of Hormuz until recent weeks; the supplied article does not place that traffic interruption in September. The Trump administration announced the U.S.-Russian diesel agreement on Oct. 9. IIR measured the global outage on Oct. 23, and the article reported that diesel futures had moved little from the announcement date.
After declining to implement his own diesel-export ban, President Trump reached a surprise agreement with President Vladimir Putin to lift sanctions and allow hundreds of thousands of tons of Russian diesel exports. The Wall Street Journal characterized the accord as a major U.S. foreign-policy shift. Energy analysts said its market effect would probably be minimal because Russia’s remaining refineries were already running at full capacity. If Russia exported the amount it had promised, the fuel would move from Russia’s domestic market rather than increase global supplies.
The available evidence supports four possible mechanisms.
First, physical-capacity displacement would occur if additional Russian exports came from Russia’s domestic market rather than increasing world supply. Full utilization of Russia’s remaining refineries is a necessary condition for that mechanism. Energy analysts cited in the article reported that condition as present. Russia’s July export suspension and the reduction in affected capacity from 6.1 million barrels per day to 1.6 million also document a prior domestic claim on Russian refining output. Those conditions and the analysts’ stated mechanism make displacement consistent with the record, but they do not prove that Russian barrels subsequently moved.
Second, an expectations-and-durability mechanism would mean buyers and sellers discounted the agreement at announcement because they expected implementation to fail or the arrangement to be reversed. The little-changed futures price is consistent with that reading, but it does not independently establish what buyers and sellers expected.
Third, an administrative-lag mechanism would place the supply effect beyond the observed window because sanctions relief had not yet generated physical cargo movements by Oct. 23. The record contains no direct export data or implementation chronology that would close this possibility.
Fourth, a substitution-offset mechanism would occur if other suppliers reduced output as Russian barrels arrived, leaving net global supply unchanged. Rystad Energy reported that Chinese refinery runs were 12% below their prewar level, suggesting possible spare capacity. Because that figure was not a post-Oct. 9 measure, it neither corroborates nor contradicts substitution in response to the agreement.
The little-changed futures tape does not distinguish among the four mechanisms. It is consistent with physical displacement because buyers expecting no net addition might not reprice at announcement, but it is equally compatible with discounted expectations or administrative delay. The documented full utilization of Russia’s remaining refineries and the prior domestic export suspension supply the physical mechanism supporting displacement; the market observation supplies no independent confirmation.
U.S. oil executives, according to people close to the industry, did not advocate for the Russian agreement. Instead, they proposed that the administration ask China to increase refinery output. Rystad Energy’s assessment of lower Chinese runs indicated possible spare capacity that could ease the global diesel shortfall. That industry position is consistent with the capacity-constraint account, but it is not direct evidence about buyers’ or sellers’ expectations.
On the available record, physical displacement passes the necessary conditions and has the strongest supporting mechanism, but there is no conclusive evidence of the sequence. The absence of export data, domestic-price data and sanctions-implementation information is not evidence against expectations, delay or substitution. The administrative-delay mechanism remains open; substitution is neither corroborated nor contradicted for the observed window. If full utilization of Russia’s remaining refineries were absent, physical displacement would be eliminated, and the muted market response would point more directly toward expectations or administrative delay.
Costs, strategic objectives and policy capacity
Jim Krane, an energy specialist at Rice University’s Baker Institute for Public Policy, described the strategic rationale for refinery attacks in two parts: “You don’t need to kill the enemy’s entire military, you just need to deprive it of fuel,” adding, “It’s a key warfighting strategy.” That is a stated rationale, not evidence that reduced fuel supply caused subsequent attack decisions.
Refinery targeting was not unprecedented for the United States or its allies despite changes in the modalities of war. During World War II, Allied forces bombed many German oil refineries, grounding Nazi fighter aircraft and damaging Adolf Hitler’s tanks and ground troops in Europe. The article presents current attacks within that historical wartime continuum rather than as an entirely novel form of conflict.
The domestic cost to Russia would fall on its own fuel market if the diesel agreement were implemented and remaining refineries remained fully utilized. More broadly, outages distribute economic pressure through fuel prices, inflation and importing-country budgets. The supplied material does not quantify the burden on any individual importing country. China’s below-prewar refinery runs indicate a possible supply role, but no export commitment or cargo volume was announced, so no resulting benefit for Chinese producers can be quantified. The available material likewise does not establish a state or private financial benefit for any party, and it does not establish private intent behind individual strikes.
In the United States, high fuel costs were driving inflation ahead of the midterm elections. President Trump was seeking ways to lower prices as congressional Republicans faced tight races. The administration pursued several instruments, but each operated within different limits.
During a trip to India the week before publication, Secretary of State Marco Rubio spoke with Indian Foreign Minister Subrahmanyam Jaishankar about increasing refined-product exports, according to people close to the two governments. Administration officials made similar requests of Greek refiners. India’s refineries were already running near full capacity, while Greek petroleum exports represented a small share of the global market. These requests therefore sought additional exports from countries with limited immediate room to increase production.
The administration also was exploring invoking the Cold War-era Defense Production Act to spend up to $150 million on additional U.S. refining capacity. The proposal remained exploratory. Any investment could take years to produce fuel because refineries are complex facilities and require logistics for supplying crude. Typical major upgrades at large refineries cost multiple billions of dollars. These instruments could influence expectations and allocation, but they could not restore the existing offline-capacity stock within days or weeks.
Feedback loops shape the recovery timeline
Fuel-deprivation loop. IIR’s attribution of outages to attack damage in Russia, Saudi Arabia, Kuwait and the United Arab Emirates supports the directional relationship between the attack rate and the offline-capacity stock. The reductions in Russian operating capacity, the reported 25% decline in Saudi refining capacity and the Middle Eastern reduction of more than 1 million barrels per day support the relationship between offline capacity and fuel availability. The subsequent links are less directly established. Reduced supply may raise the incentive and political support for continued strikes, and that support may raise the attack rate, but the supplied record documents neither the size of those feedback effects nor the subsequent decisions. The loop is therefore reinforcing by structure, with delays on the links from scarcity to political support and from authorization to launch.
Restoration loop. Repair and logistics capacity reduce the offline stock, a direction supported by Cahill’s explanation of refinery complexity. The offline stock reduces available supply, as the regional capacity figures demonstrate. Scarcity may raise the priority and resourcing assigned to repairs, but the report contains no such decision. IIR’s months-long repair expectation and the administration’s exploration of additional capacity indicate why near-term restoration remains constrained. The loop is balancing and delayed, with low near-term power because damaged plants require physical work before output returns.
Political-pressure loop. Fuel prices and inflation raise political pressure ahead of the midterms, and the administration’s policy measures have accompanied that pressure. The record documents allocation decisions, export requests and possible domestic-capacity spending but no observed price effect from those measures. The loop is balancing in direction, although its damping effect appears limited because the measures primarily address allocation and expectations rather than the physical offline-capacity stock. The Russian agreement operates within this loop as an allocation instrument; it does not remove damaged refineries.
Substitution loop. Spare capacity outside the damaged region can raise global supply, although that relationship is inferred from the capacity figures. Higher supply would narrow the global diesel shortfall, a relationship consistent with Rystad Energy’s estimate, and the shortfall induces higher refinery runs, an inferred response. The loop is balancing but narrow. India is running near full capacity, Greek exports are small, and China is the principal channel through which additional runs could occur.
On a short horizon of days and weeks, prices are governed more by political pressure, substitution, allocation and the futures market than by restoration. Read as a structural model rather than an empirical finding, repricing would be more responsive to Hormuz traffic and Chinese run-rate increases than to repair announcements. White House spokeswoman Taylor Rogers said U.S. military escorts had recently allowed more oil traffic through the Persian Gulf than before the conflict, providing a balancing channel against the earlier Hormuz disruption.
Over months to a year, restoration becomes more consequential for physical availability while continued attacks add capacity to the offline stock. The source supports that timing distinction through its months-long repair horizon and its separation of pipeline restoration from refinery work. The available snapshot does not empirically demonstrate which loop dominates.
Over years, new-capacity decisions become more consequential because an exploratory Defense Production Act investment could take years to produce fuel, while major refinery upgrades cost multiple billions and require crude-supply logistics. Ed Chow, a nonresident senior fellow at the Center for Strategic and International Studies, said, “Someone asked me if Trump was throwing a Hail Mary pass. My reaction is, he’s throwing the ball way out of bounds and saying, ‘Well, at least I tried.’”
A further escalation would connect capacity loss, fuel prices, inflation pressure and additional attacks. President Trump suggested that the United States could resume attacks on Iran after the midterms. David Goldwyn, head of Goldwyn Global Strategies and a former Obama administration official, said Iran would likely escalate strikes on oil and gas infrastructure in response. “It doesn’t look good in the Gulf,” Goldwyn said. “There could be decreased output and longer outages depending on the level of damages.” Those future decisions were not yet documented.
In Senge’s systems-dynamics vocabulary, Escalation fits the short-horizon attack-price-pressure loop. Shifting the Burden fits the medium-horizon movement of costs toward fuel supply, importing-country budgets and inflation. Limits to Growth fits the long-horizon ceiling on substitution, although the ceiling limits substitution rather than establishing an absolute absence of spare capacity. Growth and Underinvestment is visible in the administration’s stated account of years of U.S. refinery closures. Eroding Goals and Fixes That Fail capture the mismatch between an electoral price objective and a repair cycle measured in months and years. Success to the Successful cannot be tested because the record contains no capital-flow data. Tragedy of the Commons is a weak fit because the documented constraint is physical damage and engineering complexity, although Russia’s export suspension, the declined U.S. export ban and requests to India and Greece redistribute scarcity among buyers.
Public framing and evidentiary symmetry
The White House’s public account of high fuel prices was narrower than the damage distribution represented in IIR’s measure. President Trump repeatedly blamed Ukraine’s attacks on Russian oil refineries while giving less public emphasis to the Middle Eastern conflict that halted oil-tanker traffic through the Strait of Hormuz until recent weeks.
White House spokeswoman Taylor Rogers said, “As the President has said many times, today’s high oil prices reflect pre-existing issues: years of Democrat policies that shuttered U.S. refineries, compounded by the destruction of refineries in the Russia-Ukraine war—a conflict he inherited.” Those propositions and IIR’s multicausal outage measure are not mutually exclusive, but the administration’s emphasis fell on one component of the broader disruption. The article did not quantify how much U.S. refinery closures contributed to current prices.
IIR attributed the Russian refinery strikes to Ukrainian forces. The article also described drones launched from Ukraine, long-range missiles fired from Iran and Yemen, and the Houthi strike on Jazan, while describing the wider theater as a U.S.-Iran conflict.
International bodies have accused Russia, which invaded Ukraine in 2022, of intentionally targeting civilians and energy infrastructure. Ukraine has focused its strikes on military and economic assets. The Wall Street Journal reported that Moscow recently attacked Ukraine’s electrical grid, destroying 28 critical substations and leaving 550,000 households without power ahead of winter. That grid damage falls outside the nearly 12% refining-capacity measure and does not establish reciprocal refinery losses.
The same evidentiary standard applies in both theaters. IIR’s Saudi estimate combines attack damage with logistical bottlenecks, while its Russian figures combine plant losses with a domestic export suspension. The source documents attacks on electrical grids and refining assets without establishing the intent behind individual strikes in either theater.
Competing readings and the evidence that would distinguish them
Four readings remain consistent with the available evidence.
Kinetic damage dominates. The direct evidence is concentrated in the documented Russian and Middle Eastern refinery losses. This reading explains the immediate increase in offline capacity but does not determine restart timing.
Logistics and related infrastructure account for a substantial share. Pipeline and associated-infrastructure damage reduced capacity by as much as 75% at some facilities; IIR attributed Saudi Arabia’s roughly 25% decline to attacks and logistical bottlenecks; and Hormuz tanker traffic was halted until recent weeks. The report contains no pre-conflict and post-conflict throughput comparison, leaving the logistics share unmeasured.
Substitutable idle capacity is the binding constraint. Russia’s remaining refineries were already running at full capacity, while Chinese runs below prewar levels suggest possible but unquantified slack. India is near full capacity, Greece is a small supplier, and new U.S. capacity would be costly and slow. That combination limits the speed with which restoration or substitution can narrow the diesel shortfall.
Restart capability and sequencing govern duration. A moderately damaged plant could return sooner than a heavily damaged one, while complex refinery repairs take months. Without plant-level restart dates, the timing of the wider offline stock cannot be inferred from aggregate damage totals.
The most discriminating missing evidence would be a plant-level outage ledger identifying each offline unit’s cause—attack damage, scheduled maintenance or process failure—together with its nameplate capacity, current run rate and restart date. The supplied article does not provide that ledger.
If most of the 12% consisted of scheduled maintenance with near-term October restart dates, the supply-schedule explanation would carry more weight. If most consisted of damaged units without restart dates, the months-long repair constraint and scarcity of substitutable capacity would carry more weight. Exportable-capacity estimates for China, India and Greece would require nameplate capacity minus run rates, together with export and port logistics. Strait of Hormuz transit volumes before and after the traffic halt would help size the infrastructure effect.
Evidence boundaries and unresolved questions
The immediate system includes refining assets and the attacked national systems; pipelines, ports and tanker routes; spare-capacity systems in China, India and Greece; U.S. instruments actually under consideration; and the midterm calendar as the political clock.
Upstream crude production and export policy fall outside the immediate model because the supplied report contains no production data. The record therefore cannot establish whether nominally available refining capacity lacks crude. A crude-side constraint would alter the spare-capacity assessment and requires a separate test.
Demand and retail pricing also fall outside the immediate outage-composition model because the evidence concerns installed capacity and product movement rather than demand elasticity. Demand adjustments would not change the composition of the offline-capacity measure.
Private motive falls outside the evidentiary boundary because the account documents conduct, sequence and stated positions without establishing intent behind individual strikes by any party. A strategic petroleum reserve release would be a stock-side intervention rather than a capacity repair, and the report supplies neither quantities nor legal authority for such an action.
The record lacks inventory and days-of-supply data, so Rystad Energy’s approximately 2-million-barrel-per-day diesel shortfall cannot be placed against buffer stocks or classified as a physical shortage rather than a price event. Post-Oct. 9 Russian export volumes, Russian domestic diesel prices, sanctions-implementation status, buyer and seller expectations and dated Chinese refinery-run data after the agreement also remain unavailable.
IIR’s aggregation method, the country-level allocation of the global 12%, the geography of the maintenance outages and the treatment of Ukrainian refining remain undisclosed in the supplied material. That silence does not establish that no public method or dataset exists.
Plant-level restart dates and outage causes remain unavailable. A time series of outages, refinery runs, repair completions, prices, tanker traffic and policy responses would be required to establish feedback magnitudes and loop dominance. Pre- and post-conflict Hormuz transit data would be needed to size the logistics channel. Exportable capacity, port constraints and crude availability for China, India and Greece also remain insufficiently specified.
Capital-flow data would be needed to test whether investment is concentrating among incumbent operators. Upstream production data would be needed to test crude constraints, and quantities and legal authority would be needed to assess a reserve release. Independent corroboration of the IIR, Rystad Energy, futures-market, Defense Production Act and electrical-grid figures is not established in the supplied material. The supplied article remains the source of record for those figures; external verification is unresolved rather than affirmatively contradicted.
Analytical techniques used in this piece
This analysis applies the methods below. Each links to a short, plain-English explainer you can read and reuse.
- Process Tracing
- Reconstructs the step-by-step causal pathway of a specific historical event.
- Systems Dynamics (Structural)
- Maps a system’s structure — stocks, flows, and the architecture that shapes its behavior.