Charted: EU Imports 58% of Its Energy — Japan 88% — While the US Exports

Energy debates collapse three different maps into one slogan. How a country burns fuel at home is not the same as how it generates electricity, and neither is the same as who sells the molecules that cross borders. The European Union imports about 58% of the energy it uses. Japan’s dependence sits near 88%. The United States is a net energy exporter. China still runs a coal-majority primary system (~55%) while importing a rising share of oil and gas. Those facts live in separate PDFs — Energy Institute balances, Eurostat dependency tables, Ember electricity shares, GIIGNL LNG ledgers — so markets and politics often argue past each other.

This post asks the theme question directly: how do countries source, mix, and trade energy? The dashboard above is a systems ledger for 15 large economies (plus a world row in the data module). It stacks primary-energy mixes, the gap between primary and power-sector fossil shares, net import dependence, and the export concentration of LNG, crude, coal, and pipeline gas. Pair it with our global electricity generation mix for the TWh detail, and with IRENA’s 2024 renewable capacity record for the GW buildout that is reshaping future mixes — capacity additions and today’s primary balance are related, not identical.

Three ledgers, one system

Treat energy as three ledgers that only sometimes move together:

1. Primary energy mixoil, gas, coal, nuclear, hydro, and other renewables as shares of total primary energy supply (TPES). This is the “what the economy digests” meter.
2. Electricity mixhow generators produce TWh. Nuclear and hydro can clean the power stack while transport and industry still burn oil and gas in the primary stack.
3. Trade stancenet energy import dependence (imports as a share of gross available energy) and the identity of fuel exporters. A country can be coal-heavy at home and still import oil; another can export LNG while burning coal for power.

The dashboard’s Primary mix panel is the first ledger. Primary ↔ power is the second. Import map and Dependence rank are the third. Fuel trade shows who sits on the other side of the import bill.

A cross-system scoreboard

Selected 2024e figures (primary shares approximate Energy Institute Statistical Review 2025; electricity fossil shares Ember/OWID-aligned; EU import dependence Eurostat 2023):

Read the table as geometry, not morality. Japan and Korea are rich, efficient, and structurally import-dependent because they lack domestic hydrocarbons at scale. The US looks “fossil heavy” on primary shares and still exports because the surplus is oil and gas production relative to domestic demand. France imports oil and gas like other Europeans, yet its electricity fossil share collapses because nuclear covers roughly two-thirds of generation — a gap the slope panel is built to show.

Primary mix is still a fossil story

Filter the dashboard to Asia-Pacific and the stacked bars tell a blunt industrial story. China and India remain coal-majority primary systems (~55% coal each in this ledger). The United States is an oil-and-gas system (~70% combined) with coal as a residual. Saudi Arabia is oil-and-gas almost by definition. Brazil is the structural outlier among large economies: hydro and biofuels keep fossil primary near half, and electricity is already mostly renewable.

The world aggregate still sits near ~80% fossil in primary energy. That is why renewable capacity records and fossil primary persistence can both be true in the same year. IRENA’s 585 GW renewable additions in 2024 change the flow of power capacity; they have not yet rewritten the stock of oil in transport, gas in heat, and coal in Asian industry. Primary energy is a slow-moving balance sheet.

Why electricity can look cleaner than the economy

The Primary ↔ power panel plots fossil share of primary energy as bars and fossil share of electricity as a line. Three patterns matter:

• France: primary fossil near ~48%, electricity fossil near ~8%, nuclear ~64% of power. The economy still needs diesel and jet fuel; the grid does not.
• Brazil: primary fossil ~53%, electricity fossil ~12%, thanks to hydro dominance. Transport oil keeps the primary stack dirtier than the socket.
• Japan / Korea / Australia: electricity fossil shares remain high even where renewables grow, because gas and coal still firm large shares of generation while oil dominates transport.

This is the trap in “grid decarbonization = energy transition.” Cleaning TWh is necessary and measurable — see our generation-mix map — but it understates oil’s grip on primary energy until vehicles, aviation, and feedstocks move. It also overstates progress in countries that export coal or LNG while greening domestic power: Australia’s electricity renewables share can rise while the country remains a hard-coal export giant.

Import dependence is the political meter

Eurostat’s EU energy import dependency — about 58% in the latest full release used here — is the number European politics actually feels when gas prices spike. Japan (~88%) and Korea (~82%) sit higher still. Germany (~65%) is more exposed than France (~45%) even though both are EU members, because nuclear cuts France’s need for imported fossil electricity fuels even as oil imports remain.

Flip the sign and the map rearranges. The United States posts a modest net export position (about −8% in this rounded ledger) after the shale decade turned the country from oil importer to oil-and-gas exporter. Canada, Russia, Australia, and Saudi Arabia show large negative dependence — export surpluses relative to domestic TPES. That does not make their domestic mixes clean; it makes their trade stance the opposite of Japan’s.

The Import map scatter puts dependence on the x-axis and fossil primary share on the y-axis. Upper-right is the uncomfortable quadrant: high fossil intensity and high import bills (Japan, Korea, Germany). Upper-left is fossil-heavy but export-rich (Saudi, Russia, Australia). Lower-right is rarer: import-dependent systems that have already cut primary fossil intensity — France leans that way relative to peers because nuclear substitutes in the power sector even while oil imports persist.

Who sells the fuels that cross borders

Domestic mix charts hide the counterparties. The Fuel trade panel summarizes export concentration:

LNG is a triopoly story: the United States, Australia, and Qatar together clear about three-fifths of export volumes. That is why our earlier LNG capacity framing still matters for European and Asian importers even after US cargoes rewired Atlantic trade. Coal trade is more concentrated than crude: Indonesia alone is about a third of seaborne export tons in this ledger, and the top three clear roughly three-quarters. Crude is plural relative to coal — no single exporter matches Indonesia’s coal grip — yet Middle East and Russian barrels still set the geopolitical weather.

Pipeline gas remains a route-and-contract market more than a spot market. Russia’s share of inter-regional pipeline trade fell from its pre-2022 European peak, but the panel still flags Russia–Norway–Canada as the volume core. Europe’s post-2022 lesson was not “gas disappeared”; it was that pipeline dependence and LNG flexibility are different risk instruments.

What would change the story

Several observables would force a rewrite of this systems ledger:

1. EU import dependence falling under 45% for two consecutive Eurostat years without a recession-driven demand crashevidence that efficiency, domestic renewables, and diversified supply permanently cut the bill.
2. Japan or Korea electricity fossil share under 40% while primary oil share also declinesproof the socket cleanup is reaching transport and industry, not only displacing coal/gas in power.
3. China coal share of primary under 45% with oil import dependence not explodinga true primary transition, not a power-only story.
4. LNG top-3 export share under 50% as new Atlantic and Middle East trains diversify beyond the US–Australia–Qatar triangle.
5. US returning to sustained net energy importer statusa shale or demand regime change that would rewrite Atlantic LNG and oil politics overnight.
6. Nuclear and hydro additions large enough to open France/Brazil-style primary-vs-power gaps in more of Asiasee also who is pouring reactors now.

Until several of those print together, the default forecast is not “fossil disappears.” It is regional specialization: coal-heavy industrializers, oil-and-gas exporters, nuclear/hydro clean-power islands, and import-dependent manufacturing powers buying molecules from a short list of sellers.

Caveats and methodology

• Primary ≠ electricity ≠ final energy. TPES shares include transformation losses; electricity shares are generation only. Do not average them.
• Import dependence definitions differ. EU figures follow Eurostat energy-dependency methodology (2023 full release). Non-EU figures are rounded national-balance approximations aligned to IEA/EI concepts; treat them as ranks and magnitudes, not audit-grade percentages.
• Negative dependence means net export surplus relative to domestic supplyextreme values (Australia, Saudi Arabia) reflect large export industries, not “negative consumption.”
• 2024 primary shares are approximate EI Statistical Review 2025 country aggregates rounded for dashboard use; country statistical offices can differ on bioenergy and non-hydro renewables classification.
• LNG / coal / crude trade shares are secondary volume estimates for 2024 trade years; contract vs spot and re-exports can move annual ranks.
• World row is excluded from filtered panels by default because global trade nets near zero by construction.
• This post is a synthesis. For power-only detail use the generation mix post; for renewable capacity additions use IRENA 2024; for demand spikes on local grids see US data-center power vs grid capacity.

The shareable takeaway

Countries do not have one energy map — they have three. Primary mixes are still mostly fossil. Electricity can diverge sharply where nuclear or hydro dominate (France, Brazil). Trade stance separates Japan’s 88% import dependence and the EU’s ~58% from US net exports and coal/LNG surplus nations. And the fuels that move across oceans are sold by short lists: LNG top-3 ~61%, coal top-3 ~72%. Source, mix, and trade are one system — read them together or misread the transition.