Europe has 31 EVs per fast-charging port. America has 72

Paren publishes the most detailed public scorecard of North American fast charging we know of. Its Q3 2026 edition came out on 13 October. We track the European market, and we also keep our own copy of the US and Canadian charger inventories. So we put the two continents side by side and asked six questions: where they're alike, where the US sits on the adoption curve, how many cars share each fast charger, how close anyone is to break-even, what drivers pay, and how much more building would pay off.

In short, both sides of the Atlantic sell drivers the same thing, but they've built for it in opposite ways. Europe put its fast chargers in ahead of the cars. America built behind the cars, and its chargers get more use as a result. Neither approach has reached a comfortable profit yet, and the next three years will look very different on each side.

The short version

• Europe has far more fast chargers per car. The EU27 has 31 battery-electric cars (BEVs) for every DC fast-charging port. The US has 72 and Canada 67.
• On adoption, the US is roughly two years behind the EU, and California is level with Germany. At the end of 2025, BEVs made up 1.9% of US cars, which is where the EU27 stood between the end of 2023 and the end of 2024. California’s 4.8% sits between Germany’s 4.1% and the UK’s 5.0%. Norway, at 32%, is in a league of its own.
• American chargers get more use. In September a US port handled 7.4 sessions a day on average, according to Paren. In the five European markets where our usage data passes the plausibility screen, it was 5.6 over the same weeks.
• On average, no market is clearly past break-even. The industry treats 15–20% time utilization as the line. The US average of 15.9% sits at the bottom of that band. Only Hawaii, Washington DC and California clear Paren’s 20%, and our screened European markets average 10.3%.
• Europe charges more for each kWh. Before tax, the median price at a 150 kW+ charger is €0.66 in Germany and €0.64 in the Netherlands. The US average is $0.540, or €0.48, and Canada’s is €0.31.
• How much more to build (our model): if the US keeps adding ports at its 2025 pace, utilization stays at 15–21% to 2030. France, Sweden and Norway would reach 15% by 2028 with few or no new ports. Of the markets we can measure, the Netherlands is the only European one where building more fast chargers pays now. Canada, after its 2025 sales drop, is now building ahead of its cars, Europe-style.

The full story

Same product, opposite build

Start with what’s the same. Drivers on both sides pay roughly the same before tax, between about €0.45 and €0.75 per kWh. Demand clusters in the same way too. The average network is quiet while a few hotspots run hot. In the US the top 30 metro areas run at 21.1% utilization against 11.0% everywhere else, and Los Angeles reaches 31.8% (Paren, US Q3 2026). In Europe, Fastned’s Dutch ports were occupied 18.1% of the time in our four-week window, while the national average was 13.1%.

The differences go deeper. Tesla owns 49.3% of US fast-charging ports (Paren, US Q3 2026), against about 7% in the EU27 where our feeds name the operator. American sites are bigger, with 5.0 ports per site against 3.1 in the EU27, and new non-Tesla US sites average 5.0 ports too. US ports are also more powerful: 79% are rated 150 kW or more, against 57% in the EU27. The plugs differ as well, with NACS and CCS sharing the US market while Europe uses CCS almost everywhere. But the biggest difference is how many fast chargers each market has built for every car.

America is where Europe was in 2023

The measure that matters for charging demand is how many of the cars on the road are electric, not this year’s sales share. At the end of 2025 BEVs were 1.9% of the US car fleet. The EU27 passed that point during 2024, Germany in 2022 and Norway in 2015. California, at 4.8%, has as many electric cars per head of fleet as Germany (4.1%) and France (3.9%), while Texas, at 1.45%, is closer to Germany in 2021.

The sales side tells a less comfortable story. US BEV sales share was 7.6% in 2023, 8.2% in 2024 and 7.8% in 2025, so it has stopped rising. The federal clean vehicle credit ended for cars bought after 30 September 2025. In Canada the share fell from 10.9% to 6.2% in 2025 (Statistics Canada). The EU27 kept climbing to 17.4%. On the fleet the US is about two years behind the EU. On new sales the gap is closer to four years, and widening.

31 cars per fast port against 72

The EU27 has 31 BEVs for every fast-charging port. The US has 72. That gap is a policy choice. Europe’s Alternative Fuels Infrastructure Regulation (AFIR) requires a charging pool of at least 400 kW every 60 km of the core motorway network by the end of 2025, plus 1.3 kW of public charging power for every registered BEV. The rules are about coverage first and car numbers second. Germany’s federally tendered Deutschlandnetz follows the same idea.

America didn’t build that way. NEVI, its federal corridor programme, added 497 ports in 2025 (Paren, US Full Year 2025), less than 3% of that year’s new ports by our count. Most of the network is commercial, and half of it is Tesla’s. Commercial builders follow the cars, so the US ratio is lean, and leanest in its EV states: California has 94 BEVs per port and Washington 98.

Inside Europe the range is wide. Spain (12) and Italy (19) built well ahead of their cars, while the Netherlands (97) and Denmark (71) rely on dense AC networks. Norway is the interesting one. It’s the most mature EV market in the world, and it has 60 BEVs per fast port, close to the US figure. As the cars catch up with the chargers, Europe’s ratios will drift towards America’s.

More cars per port, more sessions per port

US ports averaged 7.4 sessions a day in September (Paren, US Q3 2026), which is right inside the 200–220 sessions per port per month Paren calls equilibrium. Its national utilization has stayed between 15.7% and 16.5% for five quarters while the port count grew 26% in a year. Sessions grew 26% too. Every new port has been filled at the same rate the cars arrived.

Our five screened European markets averaged 5.6 sessions a day over the same four weeks, with ports occupied 10.3% of the time. The Netherlands, at 97 BEVs per port, matches the US at 7.6. Norway (60) gets 6.7 and France (32) gets 4.9. Fewer ports per car means more cars per port, as you’d expect. Canada’s 4.4 says the same from the other side. Paren reports Canadian sessions up 37% in a year while the tracked port base grew 51% ({pca()}).

Break-even is a band, not a line

Everyone quotes a break-even number, but they don’t mean the same thing by it. Paren draws its “Healthy” line at 20% utilization. Daniel Breton of Electric Mobility Canada, in the guest essay of Paren’s Canada Q3 2026 report, calls 20% “the threshold that the industry needs to be financially healthy”. Paren’s 200–220 sessions per port per month is a different idea, a balance between drivers waiting and capital sitting idle, not a profit line.

The European number is 15%, from the EV Charging Masterplan that ACEA and seven other industry associations published in 2022 with McKinsey’s help. It says 15% average utilisation “is required for economic viability in the medium to long term”. Its utilisation is measured in energy, though, not time. In our deep dive on faster charging we estimated that cars on chargers rated 269 kW on average draw 59–66 kW while plugged in. At that rate a 15% energy share would need the port occupied 60–70% of the time, which is queue territory in our data. That conversion is our own calculation, but it means the 2022 figure fits a 50 kW network better than today’s 300 kW one.

In practice, break-even depends on the site. Using German data from 2021, Hecht, Figgener and Sauer (iScience, 2022) found that 59% of 100–200 kW charge points covered their costs from electricity sales at an assumed margin of €0.50 per kWh. For chargers of 200 kW and more the figure was 29%. EVgo, the listed US fast-charging operator, sold 276 kWh per stall per day in Q2 2026, 2% less than a year earlier, while its stall count grew 24%. Its throughput per stall has stayed flat while the network grew, which matches what Paren sees nationally.

Measured against a 15–20% band of time utilization, the US average sits at the bottom. Only three places clear Paren’s 20% line: Hawaii (32.4%), Washington DC (27.5%) and California (23.4%) (Paren, US Q3 2026). Our screened European markets average 10.3%, with the best networks in the mid-teens: Fastned in the Netherlands at 18.1% and Circle K in Norway and Tesla in France at 13.9%. That’s a lot of expensive hardware still waiting for cars. Europe’s regulation was written expecting exactly that.

Europe charges more for each kWh

Before tax, a 150 kW+ charger costs a median €0.66 per kWh in Germany, €0.64 in the Netherlands and £0.63 (€0.74) in the UK. The US average is $0.540, or €0.48. Outside Tesla it has sat at about $0.533 for five quarters, and all of the national increase came from Tesla, whose non-member price rose 15% in a year to $0.569 (Paren, US Q3 2026). Canada is cheapest at CAD 0.486 (€0.31). Spain, at €0.46, is the one European market priced like the US.

So European operators sell fewer kWh per port at a higher price, and American operators sell more at a lower one. A German 150 kW+ charge costs about 37% more per kWh before tax than the US average, which makes up for a good part of the gap in use. We can’t yet put the two sides into one revenue-per-port figure. That would need energy per session on both sides, measured the same way, and we don’t have it.

How much more building pays

Start with the US. If it keeps adding about 19 800 ports a year, as it did in 2025, utilization in our model sits at 15.0–17.2% in 2028 and 15.8–20.9% in 2030, across our low and central adoption cases. In other words, America is building at about the rate that keeps its chargers exactly as busy as they are now, which is what five flat quarters in Paren’s reports have shown. In our high case it reaches 25.6% by 2030, enough to bring queues to the metros already above 20%.

Canada has gone the other way since its sales fell. BEV share of new cars dropped from 10.9% to 6.2% in 2025, but it kept building: 1 376 new ports from January to September 2026 (Paren, Canada Q3 2026). Our model says 3–28% more ports than today would hold Canadian utilization at 15% by the end of 2028. At its 2026 pace Canada will have about 43% more, which puts utilization at 11–14%. Canada is now building ahead of its cars, Europe-style.

Europe is the other way round. If France, Sweden and Norway stopped building today, their utilization would still rise to about 15% by 2028: 15.6–18.9% in France, 15.3–17.2% in Sweden and 14.1% in Norway. To land at 15% at the end of 2028, France has room for roughly 2 000 to 13 000 more ports, Sweden up to about 2 000, and Norway none. The Netherlands needs 32–47% more fast ports by 2028 to stay at 15%. It’s the only European market we measure where more fast chargers are clearly needed now.

That doesn’t make AFIR’s corridor ports a mistake. Coverage is a policy goal, and a car needs a charger every 60 km whether or not that charger pays its way this year. But a CPO choosing where to put the next euro faces a different calculation from a regulator. On today’s data, Europe’s best fast-charging investment for the next two or three years is getting more use out of the ports it already has.

What we’re watching

Three things would change this picture. US sales share has been flat at about 8% for three years, and without the tax credit the low case is the one to take seriously. In Europe, mature markets like Norway show that ratios converge: once the cars arrive, the 31 BEVs per port moves towards 60, and utilization will rise whether anyone plans for it or not. And Paren’s 2026 “slowdown” is largely a comparison of new counts with old ones that have since been revised up. We’ll keep comparing each quarter’s first-published figure with the next one.

Our data shows the same thing as Paren’s. America’s chargers have to keep up with the cars, and Europe’s cars are still catching up with the chargers.

Sources