Europe’s Energy Map Was Already Changing

In 2021, seven major European importers -- Germany, Italy, France, the Netherlands, Poland, Spain, and the United Kingdom -- bought about $431 billion of crude oil, refined products, and natural gas, measured as bilateral goods trade. The Russian Federation was the single largest source, supplying about 18.6% of that combined energy bill.

By 2024, Russia’s measured source share across the same seven importers had fallen to 1.8%. In absolute terms the drop is starker still: Russian-origin energy into the group fell from roughly $80.3 billion in 2021 to about $9.1 billion in 2024, even though the group’s total energy imports were higher in nominal terms. The replacement supply is visible in the same record — the United States rose from about $33.9 billion to $82.3 billion, and Norway held its position as the largest single supplier at around $76.9 billion.

This article reads that realignment through one specific, measurable lens: the source country of each importer’s energy, by product. That is a fact in the trade record. It is deliberately not a claim about the route any cargo took, nor about the exact month the change happened. The data is annual, so the 2022 shock appears as a step between years, not a dated event. We keep the measured origin shares and the external policy timeline strictly apart throughout.

Germany is the clearest case, and it complicates the simple story. Russia was indeed a leading energy source for Germany before 2022: in 2018, Russia supplied about 23.5% of Germany’s in-scope energy imports — the largest single share that year — and it held near 21% through 2020. But the share was already easing to about 15.0% by 2021, the last full year before the invasion. The title of this article is literal: the map was changing before the shock, not only because of it.

What 2022 did was turn a gradual decline into a near-total cutoff. Germany’s Russian energy source share fell to 0.7% in 2023 and to effectively 0.0% (0.01%) in 2024. By then Germany’s largest measured sources were Norway (27.1%), the Netherlands (14.5%), and the United States (13.7%).

Two cautions belong right next to these numbers. First, the Netherlands and Belgium appear as large 'sources’ for Germany, but both host major re-export hubs (Rotterdam, Antwerp), so part of that share is energy transiting Europe rather than originating there — source country is not always the original producer. Second, Germany’s total energy bill swung from about $88.8 billion in 2021 to $174.8 billion in 2022 before falling back; most of that spike is price, not volume. The share shift is robust; the dollar levels must be read with the 2022 price surge in mind.

Pulling the seven importers together and comparing 2021 with 2024 shows the substitution at the level of origin countries. The largest single change is the United States, whose measured energy deliveries to the group rose from about $33.9 billion to $82.3 billion -- a jump that moved it from a mid-tier supplier to a rival of the long-standing leader. Norway remained the anchor of European supply at roughly $76.9 billion, essentially flat in value but far more prominent in share once Russia receded.

Beyond those two, the gains are spread across many origins rather than concentrated in one: Algeria (about $19.1B to $26.5B), Saudi Arabia ($16.6B to $25.2B), and Kazakhstan ($11.3B to $19.4B) all rose, while Russia fell from about $80.3 billion to $9.1 billion. The diversification is the point: no single new supplier replaced Russia one-for-one.

The accompanying map is schematic. It links measured source countries to importing countries to orient the reader; it does not trace pipelines, shipping lanes, or vessel routes, and the dataset cannot observe any of those. Endpoints are an origin country and a destination country in the trade ledger — nothing more. We use no directional or conflict imagery; the map is a reference for who supplied whom, by value, not a depiction of how energy physically moved.

Splitting natural gas into its two product codes is where the realignment is most vivid — and where the limits of the data must be stated most carefully. HS6 271111 (liquefied) and 271121 (gaseous state) are product-state codes, not infrastructure. 'Gaseous’ is a reasonable proxy for pipeline-type trade and 'liquefied’ for LNG, but neither code identifies a pipeline, a terminal, or a route.

With that caveat, the contrast is striking. For the seven importers, measured LNG imports jumped from about $27.9 billion in 2021 to a peak of $105.9 billion in 2022, then settled at $37.4 billion in 2024 -- still well above pre-war levels. Inside that LNG total, the United States' share rose from 27.9% to 49.0% between 2021 and 2024: by 2024 roughly half of the group’s liquefied-gas imports were US-sourced. Qatar's LNG share, by contrast, fell from about 19.9% to 7.5% as American volumes crowded the mix.

Gaseous (pipeline-type) gas tells the opposite story for Russia. Russia’s share of the group’s gaseous-gas imports fell from 12.5% in 2021 to 2.4% in 2024, while Norway supplied roughly 46% of gaseous gas in 2024. One nuance worth flagging honestly: Russia’s share of LNG was still about 16.9% in 2024 -- Russian liquefied gas kept flowing even as Russian pipeline-type gas was largely displaced, a distinction that a single 'gas’ aggregate would erase.

The aggregate hides how differently each importer re-sourced. Poland shows the most dramatic break: Russia supplied about 51.2% of its in-scope energy in 2021 and effectively 0.0% in 2024. Its 2024 mix leans on Saudi Arabia (33.3%), Norway (20.5%), and the United States (16.0%) -- a wholesale reorientation away from a single dominant supplier.

Germany went from 15.0% Russian in 2021 to near zero, replacing it mainly with Norwegian gas (27.1%) and rising US volumes (13.7%), though intra-EU hub flows (Netherlands 14.5%, Belgium 11.1%) inflate the apparent role of neighboring countries. The Netherlands -- itself a trading hub — moved from 27.9% Russian to 1.5%, with the United States becoming its largest source at 20.3%. Italy cut Russia from 24.3% to 3.0%, but diversified toward its own neighborhood rather than the Atlantic: Algeria (15.0%), Azerbaijan (11.9%), and Libya (10.3%) lead its 2024 mix.

The common thread is a fall in Russian share to low single digits or zero; the differences are in the destinations. Northern importers leaned on Norway and US LNG; Mediterranean importers leaned on North African and Caspian pipelines and nearby suppliers. Each card reports source country as measured, not route, and intra-EU rows should be read as partly transit.

Two limits are central enough to state on their own, not bury in a footnote.

First, timing. BACI is annual. The 2022 invasion, the sanctions packages that followed, and the loss of major pipeline flows all happened within a single year, but in this data they appear only as a step from 2021 to 2022 to 2023. The series can show that Russia's share fell from 18.6% to 11.4% to 1.8% across those years; it cannot locate the inflection within a year, distinguish a spring cut from an autumn one, or capture spot diversions. Any monthly narrative requires monthly data this dataset does not contain.

Second, routes. Every figure here is a source country, not a path. Gaseous-gas codes do not identify pipelines, LNG codes do not identify terminals, and intra-EU sources such as the Netherlands and Belgium partly reflect re-export through hubs rather than original production. The collapse in Russian source share is real and measured; reading it as a specific pipeline being switched off, or as proof of which sea lane carried a replacement cargo, would be an inference the data does not support.

Finally, causation and price. The trade record is consistent with the well-documented external timeline of sanctions and pipeline disruption, but it does not by itself prove that those policies caused the shift — that linkage is context, cited separately. And because values are nominal, the 2022 spike to about $798 billion for the seven importers reflects the energy price surge as much as any change in quantities.

Every value in this article is computed from CEPII BACI HS92 bilateral trade for 2018-2024, via a committed, re-runnable query. The measured layer is each European importer’s energy purchases by source country across four HS6 codes: crude petroleum (270900), refined petroleum (271000), liquefied natural gas / LNG (271111), and natural gas in gaseous state (271121). Source share is the value from a given origin divided by the importer’s total in-scope imports of those four products from all sources. The seven importers are resolved by BACI numeric code through the project dimension table (Germany 276, Italy 380, France 251, Netherlands 528, Poland 616, Spain 724, United Kingdom 826); Norway is code 579 and the United States 842.

The proxy layer is the interpretive step from product code to physical reality. 'LNG’ and 'gaseous’ are product states, not infrastructure; source country is not route; and intra-EU sources can be re-exports. We label these as proxies and never present them as route or origin-of-production measurements.

The context layer is external and editorial. The 2022 invasion, subsequent sanctions, and the loss of major pipeline flows explain why the realignment happened, but that timeline is sourced outside the trade data and is kept separate from the BACI numbers. Standard data caveats apply: BACI annual data cannot show intra-year shock timing; natural-gas codes do not identify pipeline routes; BACI numeric codes can differ from ISO numeric codes; the product history is HS92-compatible through 2024; and source-country and product-code continuity were validated before publishing. No figure here was reasoned out — each one comes from the generator script that produced the chart data.