Goldman Says Europe is Safe. Its Chart Ends Before the War.
Bloomberg / Getty Images
You’re the owner of a manufacturing facility in Europe; business is going as usual, and you lock up for the night after a long day’s work. You wake up the next morning, and on your way to work, you check the news and learn your energy bills have tripled overnight. Panic sets in, as you realize you only have two choices lying ahead of you: absorb the loss and wait out the crisis, or close your doors forever. What would you do?
In 2022, European firms faced this exact problem. On 24 February 2022, the day Russia invaded Ukraine, Dutch gas futures (TTF) rose as much as 62%, the biggest jump since 2005, while Brent went as high as $105 for the first time since 2014. By the end of 2022, European gas averaged €133/MWh, a staggering increase of 180% on 2021. In response, roughly 9% of the continent’s chemical capacity was announced for closure between 2022 and 2025, with energy costs the stated reason in about half the cases, and when prices fell in 2023, it never returned. If this wasn’t bad enough, high energy prices reduced European competitiveness, with the EU’s imports-to-production ratio in energy-intensive sectors rising by 11% in 2022. Today, with the situation in the Middle East and prices rising once again, will the continent’s firms face the same nightmare once more?
Goldman Sachs argues no. In their report on 20 April 2026, “This Time is Different (from 2022) - The Impact of Higher Energy Prices on European Manufacturing Industries”, Goldman analysts provide three main arguments. Goldman’s first argument is a distinction that less careful onlookers often miss: this is an oil shock, not a gas one, and that matters when looking at which sectors will be most impacted by the crisis. The reaction in 2022/2023 was driven by four gas-intensive industries, which accounted for about 70% of European industrial gas consumption, according to Goldman, an estimate slightly higher than Bruegel’s and CGEP's, at 65% and 66%. Gas shocks concentrate in a few energy-intensive sectors; oil shocks hit manufacturing more broadly, spreading the cost. This is because oil markets are more globally integrated than gas markets.
Second, Goldman proposes that the shock isn’t geographically asymmetric like in 2022, which was concentrated in Europe, where energy prices rose by almost 200%; Goldman puts this year’s rise at only 44%, against 66% in Asia, meaning that European goods and services shouldn’t become less competitive as they did in the last energy crisis.
Third, and this is the load-bearing one, it argues that the shock will be smaller and shorter than in 2022, citing their own internal projections that annual-average energy prices will only be about 20-30% higher than in 2025, with the crisis being less persistent than the Russian gas cut-off, which was understood to be permanent.
Goldman was not alone. In April, the IMF assumed a short-lived conflict with a 19% rise in 2026 energy prices, with Vanguard in the same period cutting euro-zone GDP growth by only 0.4 points, from 1.2% to 0.8%, conditional on elevated prices lasting one to two quarters. This was the broader consensus at the time these reports were made. The ECB, holding rates on 30 April, judged the incoming data broadly consistent with its projections; the OECD built the baseline of its June outlook on a scenario it named “time-limited disruption”, in which Gulf production and exports return to pre-conflict levels from the third quarter. Every one of these forecasts was made between April and June, and every one took the short branch.
Everything in Goldman’s sector logic rests on the first argument. If this is an oil shock, the cost spreads across manufacturing and no single industry has to make the factory owner’s choice. If it is a gas shock, the four industries that carried 2022 carry it again. So the question is not which commodity was hit first in the Gulf; it is which commodity Europe ends up paying for.
With the closure of the Strait of Hormuz, 20% of the global supply of LNG is currently blocked; LNG only makes up 15% of global gas supply, according to Goldman, and about 45% of the EU’s supply; these figures are in consensus. That last number is the one that matters. A blockage that is small for the world is large for a continent that replaced Russian pipeline gas with tankers, and the market has priced it as such: the TTF reached €81.07/MWh on September 16 after an annual high of €83.97 the day before, the highest since December 2022, an increase of 187% since the start of the year, just for it to drop 8% to €70 on the mere whispers of a possible new diplomatic resolution on Hormuz. Qatar doesn’t think the situation is going anywhere either, with it applying force majeure to postpone delivery through autumn to avoid paying penalties.
A gas price that has nearly tripled since January is not the signature of an oil shock spilling over. It is a gas shock, and it lands where 2022 landed: on chemicals, fertilisers, glass, steel and the rest of the 70% of industrial gas consumption Goldman itself identified. The cost is not spread out. It is concentrated, in the same places, again.
Goldman’s second argument compares prices: Europe up 44%, Asia up 66%, so Europe loses no ground. But a price index measures what energy costs today, not whether there will be enough of it in January. Asia buys LNG into demand. Europe buys it into a storage deadline, and this year the deadline is being missed.
European gas storage stood at 66.59% on 5 September, the lowest for that day since records began in 2011, with a target of 80% by 1 November, already relaxed from the usual 90% due to current geopolitical conditions. The European Commission, in the meantime, has raised its TTF futures assumptions by 47% for 2026 and 32% for 2027. That is the Commission telling the market it does not expect the gap to close at a normal price.
So the asymmetry is not in the price rise. It is in what each region has to do about it. A European firm is not only paying more per unit than it did in January; it is competing with its own government for cargoes that have to be in the ground by November, at whatever price fills them. No Asian competitor is running that clock. That is the competitiveness loss Goldman’s comparison cannot see.
All of these firms and institutions based the foundations of their assumptions on one variable, which careful readers may have caught: persistence. This is of particular interest for European firms, as a crisis that lasts 6 months is two bad quarters; one that lasts 10 years calls for a shutdown. Goldman understood this as well, with it stating that part of why 2022 was so disastrous was that the loss of Russian gas was understood to be permanent; in that case, the arithmetic of persistence was solved from the start.
This time, Goldman did put a number on it: prices are predicted to hit above €100/MWh if Middle East exports normalize gradually only in 2027, and €50 if operations return to normal before then. Vanguard assumed the crisis would last only one to two quarters, 90 to 180 days. The OECD had a second scenario too, “prolonged disruption”, with supply disruptions persisting well into 2027 and global growth falling to 1.8%; it was published as the downside, not the forecast. Consensus priced the €50 branch.
Right now the world is going in the opposite direction. The war has surpassed 200 days, and since the ceasefire collapsed on 8 July, the diplomatic stance has only hardened, with Iran insisting on lifting the US blockade before negotiations. To top that, Saudi supply is now heavily constrained by Houthi attacks, and while the situation in the Suez is softening, insurance isn’t, making it cheaper for many ships to circumnavigate Africa instead of paying the risk premium associated with the Red Sea. The time it takes for prices to lower depends not only on the end of war, but also on the speed of resuming regular operations in Hormuz, and on no new events creating further disruptions, such as the Houthis in the Suez. Every input has moved to the €100 branch. The ECB, which in April saw data matching its baseline, raised rates in June for the first time in nearly three years, and said the decision held across every scenario for how long the shock lasts.
And the evidence Goldman offered that this shock is symmetric, the chart that says Europe will not lose ground, is a comparison of Chinese and European petrochemical prices. Its last observation for the euro area is February 2026. Hormuz closed in March. The chart that says Europe is safe ends before the war.
But how does this affect European industry? It will lose capacity, that much is certain, even if the war ended tomorrow, things wouldn’t go back to normal immediately. But the full effect on capacity won’t be visible to us, not yet. The damage from 2022 wasn’t apparent in 2022; European manufacturing held at first, decelerating in the first quarter of 2023, stalling in the second, turning negative only after. Production in energy-intensive industries fell by 3.7% in 2022, and 12.4% by 2023 compared to 2021 levels, even though gas prices came down during the year; the industrial decline accelerated anyway. Anybody reading the European industry at the time would have had the same picture we have now, and reassuring data isn’t a sign that things will be mild. Every number cited and used by Goldman at the time was correct. The shock may be more spread out; it may be more asymmetric; it may be temporary, but temporary under what timeframe?
Goldman’s own headline number is a 2% hit to European industrial production by the end of 2027 against its pre-conflict baseline. That is an all-industry figure, and it is the shape of the shock Goldman expected: shallow, spread out, and over. Our base case is that European energy-intensive output in the first half of 2027 sits 5-7% below the same period of 2025, whether or not Hormuz reopens, because the 2022 pattern says the damage is booked in the quarters after the price peak, not during it. Chemicals go first, and within chemicals the gas-feedstock lines: ammonia, methanol, the plants that already came closest to closing in 2022 and did not get a recovery to justify reinvestment. Glass and ceramics follow, because they cannot bank capacity and restart it. What would make us wrong is simple to state and hard to see: storage at 80% by 1 November and a Hormuz timetable the market believes, both before the first-quarter budgets are set. Neither is on the board today.