A Data Investigation · June 2026

How China Won the Electric Car

From the lithium refinery to the showroom floor, the numbers behind Beijing's — and BYD's — command of the defining industry of the decade.

Compiled from 26 sources · 22 claims survived adversarial fact-checking · IEA, SEC filings, CPCA, CSIS, BloombergNEF
0
of the world's EVs
built in China, 2025
0 vs 0
BYD vs Tesla
2025 BEV sales
0
of global battery
cell production
0/kWh
China pack price —
EU pays 56% more
Scroll ↓
Before you read · Test your gut
How much do you think China leads?
Drag the slider to your guess, then reveal the answer. Three quick questions.
Chapter One · Market Dominance

The Crown Changes Hands

BYD Seal electric sedan, BYD's direct rival to the Tesla Model 3
The BYD Seal — the company's direct answer to the Tesla Model 3, and a symbol of the year the crown changed hands.
BYD's lineup now spans a $9,500 city car to executive sedans and supercars.Photo: Rutger van der Maar, CC BY 2.0, via Wikimedia Commons

For a decade, the electric car had one undisputed king. That era ended on January 2, 2026, when the full-year numbers landed: BYD sold 2,256,714 battery-electric vehicles in 2025 — up 27.9% — while Tesla delivered 1,636,129, down 8.6% in its second consecutive annual decline. The gap: more than 620,000 cars. For the first time in history, the world's best-selling pure-electric carmaker is Chinese.3,4,5 Verified 3–0

The collapse was steepest at the finish line. Tesla's Q4 2025 deliveries fell to 418,227 — down 15.6% year-over-year, missing consensus.3 And the crown was not narrowly won: just one year earlier, in 2024, Tesla had clung to the annual title by fewer than 25,000 units (1,789,226 vs 1,764,992). In twelve months, a near-tie became a rout.

Fig. 1 — The Overtake
Global battery-electric vehicle sales, BYD vs Tesla
Full calendar years. BYD: passenger BEV wholesale. Tesla: deliveries.
BYD (BEV only) Tesla
1,764,992
1,789,226
2024
2,256,714 ▲ 27.9%
1,636,129 ▼ 8.6%
2025
BEV-only comparison; BYD additionally sold ~2.3M plug-in hybrids for ~4.6M total NEVs in 2025. Sources: Tesla SEC 8-K delivery report; BYD production & sales filing; CnEVPost, Electrek, CNBC. Verified 3–0 across five constituent claims. Note: Tesla retook the quarterly BEV lead in Q1 2026 as BYD's domestic sales fell ~25% — see Chapter 8.
Fig. 1b — The Race, Animated
Global pure-electric (BEV) sales by manufacturer, 2021 → 2025
Watch the order change. Press play, scrub the years, or tap a year below.
2021
Units = battery-electric only (PHEVs excluded). Manufacturer groups: Geely Holding includes Volvo/Polestar/Zeekr; SAIC includes the Wuling JV. Firm figures for BYD, Tesla, VW, BMW; SAIC/Geely 2024–25 are estimates (±5%) derived from EV-Volumes group shares. Sources: EV-Volumes / CleanTechnica / InsideEVs league tables, company filings. The crossover — Tesla leading 2021–24, BYD seizing #1 in 2025, Geely surging from ~100k to ~1.2M — is well-supported.

At home, the rout is more lopsided still. BYD finished 2025 as China's #1 automaker overall — not just in EVs — with 14.7% of the entire passenger-car market, ahead of Geely (11.0%) and every foreign brand. In China's NEV market it held 27.2% retail share (3.48 million units). Tesla? Fifth place, with 4.9% — 625,698 units, down 4.8% — having slipped from third the year before.7 Verified 3–0

"In twelve months, a near-tie became a rout. The world's largest car market no longer has a foreign company in its top four."
The Lopsided Playing Field

BYD won the crown with one market shut to it entirely

The global sales figures understate the achievement. BYD became the world's best-selling EV maker while effectively banned from the United States — the planet's second-largest EV market. Tesla, by contrast, builds and sells freely inside China.

🚫 BYD → North America
Doors bolted shut
United States — effectively prohibited
A 100% tariff on Chinese EVs (2024, maintained 2025) plus a 25% auto tariff pushes the burden past 125%. The Connected Vehicle Rule (finalized Jan 2025) then bans Chinese-linked vehicle software from model-year 2027 and hardware by 2030 — even if the car is built on US soil. Not merely taxed: barred.
Canada — slammed shut, then cracked open
Matched the US with a 100% tariff in Oct 2024. Only in early 2026 did Ottawa ease it — to 6.1% on a capped quota of 49,000 vehicles (rising to 70,000 by 2030) in a canola-for-EVs trade deal. BYD has since confirmed a late-2026 Canadian launch.
✓ Tesla → China
Full run of the home market
A wholly-owned Shanghai gigafactory
Tesla is the only foreign automaker permitted to fully own its China plant — built in under a year in 2019 — and exports from it across Asia and Europe. No Chinese partner required, no quota.
Selling deep inside BYD's backyard
625,698
Tesla vehicles sold in China in 2025 — 4.9% of the NEV market. Down 4.8%, but a volume BYD cannot reciprocate with a single legal sale in the US.
The world's #1 EV maker is locked out of the world's #2 market — while its closest rival builds cars in the #1 market's largest city. The crown was won on an uneven field.17,18,7
Chapter Two · Production & Exports

The Factory of the World, Electrified

One company is the headline; one country is the story. Of the nearly 22 million electric cars produced worldwide in 2025, China built roughly 16 million — about 75%. Chinese factories produced 20% more EVs than Chinese buyers purchased, and the surplus poured outward: exports doubled to a record 2.5 million+ units, roughly 40% of all global EV trade.1 Verified 3–0

Fig. 2a — Production
Where the world's EVs were built, 2025
Share of ~22M electric cars produced globally (incl. PHEVs)
75% MADE IN CHINA
~16M of ~22M vehicles
IEA Global EV Outlook 2026
Fig. 2b — Conquest of the Global South
Chinese share of EV sales outside the US & Europe
Chinese imports as share of electric-car sales in the rest of the world
55% UP FROM ~10% IN 2021
5.5× share gain in four years
IEA Global EV Outlook 2026
0
Chinese EV exports in 2025 — doubled year-over-year to a record
0
Chinese-made EVs sold in Europe in 2025, up ~50% despite tariffs
0
of Chinese-made EVs sold overseas are now Chinese brands — up from <40% in 2021
0
BYD's own 2025 exports — past 1M for the first time, up ~150%
Live · since you opened this page
0
electric vehicles have rolled off Chinese assembly lines as you read — at roughly 16 million a year, about 30 every minute.

The export story marks a structural shift, not a volume blip. In 2021, Chinese-made EVs sold abroad were mostly foreign brands — Teslas and Dacias built in Shanghai. By 2025, 80% were Chinese brands. China stopped being the West's contract factory and became its competitor, with BYD alone shipping 1,046,083 vehicles overseas — up roughly 150% in a single year.1,4 Verified 3–0

Interactive — The Tariff Map
Where the world lets Chinese EVs in
Tap any market. The door is wide open across the Global South and shut hardest where the incumbents are strongest.
Status as of mid-2026. Tariff/regulatory positions from IEA GEVO 2026, CSIS, the Globe and Mail, Electrive, Atlantic Council and government announcements. Node size roughly reflects EV-market importance, position is schematic. The US figure combines a 100%+ tariff with the connected-vehicle rule; "easing" reflects Canada's early-2026 quota deal.
Interactive — It May Already Be in Your Driveway
Who owns whom: the "Western" brands that are Chinese
A surprising amount of European and British car heritage now answers to Chinese parents. Tap a brand.
Lines link Chinese parent companies to brands they own or control; stakes are approximate and widely reported as of 2025–26 — confirm against current filings. Geely Holding is the standout (Volvo, Polestar, Lotus, Lynk & Co, Zeekr, half of Smart, the London black-cab maker, plus ~17% of Aston Martin); SAIC owns MG. The Stellantis–Leapmotor tie is the reverse: a Western giant taking a stake (~20%) in a Chinese brand and selling it abroad. Sources: company filings, Geely/SAIC/Stellantis disclosures, Reuters.
Chapter Three · The Supply Chain

Owning Everything Beneath the Car

Market share can be lost in a model cycle. What China controls beneath the car cannot. The International Energy Agency now identifies China as the dominant refiner for 19 of the 20 energy minerals it tracks, with an average market share near 70% — and concentration is rising, not falling.2 Every battery chemistry, every traction motor, every cathode powder routes through Chinese industrial capacity at some point between the mine and the showroom.

The Journey — Mine to Showroom
Follow the material, and China is at every step
A lithium atom's path from the ground to your driveway — and China's share at each stage.
⛏️
~70%
Mining
rare-earth mining; major in lithium & graphite
⚗️
~70%
Refining
#1 refiner for 19 of 20 minerals
🧪
85–90%
Materials
~85% cathode · >90% anode
🔋
>80%
Cells
of the world's battery cells
🚗
~75%
The car
of all EVs built in China
🚢
1M+
Export
BYD units shipped on its own fleet
Shares as detailed in Fig. 3 and the battery panel below. The point: vertical control isn't one chokepoint — it's the whole chain, mine to ship.
Fig. 3 — The Chokepoints
China's share of the global EV supply chain, layer by layer
2025 production / processing shares. Hover any bar for sourcing.
Battery cell production>4 TWh global capacity; EU & US ~6–7% each
>80%
Cathode active materialthe most valuable part of the cell
~85%
Anode active material
>90%
Purified phosphoric acidessential for LFP batteries
75%
High-purity manganese sulphate
95%
Rare-earth mining
~70%
Rare-earth separation & processing
90%
Rare-earth magnet manufacturingthe heart of every EV traction motor
93%
Mineral refining, avg. across 19 of 20IEA-tracked energy minerals where China is #1
~70%
Sources: IEA Global EV Outlook 20261; IEA Global Critical Minerals Outlook 20252; rare-earth rows from CSIS, Oct 20258. All bars verified 3–0 by independent fact-check panels. Concentration is increasing: the top-3 refiners' average share rose from 82% (2020) to 86% (2024), with ~90% of recent supply growth coming from the single top supplier.
Fig. 3b — The Battery Oligopoly
Two Chinese companies make more than half the world's EV batteries
2025 global EV battery installations, 1,187 GWh total (+31.7% YoY). Six of the top ten makers are Chinese.
CATL 39.2%
BYD 16.4%
CALB
Gotion
LG 9.2%
Others
Chinese makers ≈ 70% Korea & Japan (LG, SK On, Panasonic, Samsung) Others
CATL + BYD alone = 55.6% of the world's EV batteries (659.5 of 1,187 GWh). BYD makes its own cells in-house and still ranks #2 globally while also building cars. Source: SNE Research via CnEVPost.19

The chemistry of the modern EV was effectively chosen in China. Lithium-iron-phosphate (LFP) batteries — cheaper, safer, cobalt-free, and pioneered commercially by BYD's Blade pack — grew from under 10% of the global EV battery market in 2020 to over 55% in 2025.1,2 The world standardized on the battery China is best at making — and China supplies 75% of the purified phosphoric acid it requires.

Nor is this a lead the West expects to erase. Even in the IEA's own forward scenarios, China still supplies ~90% of anode material, ~75% of cathode material, and two-thirds of all batteries in 2035 — a decade from now, after every announced Western gigafactory is counted.1 Verified 3–0

The cells of a BYD-style LFP Blade battery pack
Inside the Blade: the long, flat lithium-iron-phosphate cells that BYD made mainstream — now the chemistry behind a majority of the world's new EVs.
LFP grew from under 10% of EV batteries in 2020 to over 55% in 2025.Photo: BoldLuis, CC0 / Public Domain, via Wikimedia Commons
"Market share can be lost in a model cycle. The refinery layer cannot. Even the IEA's 2035 scenario leaves two-thirds of the world's batteries in Chinese hands."
Chapter Four · The Financials

The Money: Bigger Top Line, Bigger Bottom Line

BYD first out-earned Tesla on revenue in 2024 — $107 billion to $97.7 billion. In 2025 the gap widened in both directions: BYD grew to RMB 804 billion (~$116B) while Tesla posted its first-ever annual revenue decline, to $94.8 billion. And despite fighting the most brutal price war in automotive history at home, BYD's net profit of $4.7 billion still exceeded Tesla's $3.8 billion — which had fallen 46%.10,11,12

FY 2025
BYD
TESLA
RevenueBYD: RMB 803.97B converted
~$116B+3.5%
$94.8B−3%
Net incomeBYD attributable · Tesla GAAP
$4.72B−19%
$3.79B−46%
Vehicles soldBYD: all NEV · Tesla: deliveries
4,602,436+7.7%
1,636,129−9%
Automotive gross margin
20.5%
18.0% (total GAAP)
R&D spendBYD: RMB 63.4B, +17% YoY
$9.2B
Engineers on staff
>120,000
● = leader · BYD figures: FY2025 annual report (Mar 27, 2026) · Tesla figures: SEC 8-K (Jan 28, 2026) · Company filings, not adversarially fact-checked — see methodology

The structural advantage sits below the income statement. According to BloombergNEF's December 2025 survey, a battery pack in China costs $84 per kWh — North America pays 44% more, Europe 56% more. With the pack still the single most expensive component of an EV, a ~30% battery-cost advantage is a head start no Western assembly-line efficiency can claw back.1,9 Verified 3–0

Fig. 4 — The Cost Moat
Average lithium-ion battery pack price by region
BloombergNEF survey, December 2025 ($/kWh)
Chinahome of >80% of cell production
$84 /kWh
North America
≈$121  (+44%)
Europe
≈$131  (+56%)
Regional pack-price averages reflect chemistry mix (China is majority-LFP) and manufacturing scale, not identical-cell teardowns. Sources: BloombergNEF Dec 2025; corroborates IEA's ~30% Chinese battery cost advantage. Verified 3–0.
Fig. 4b — The Floor Keeps Dropping
Lithium-ion battery pack prices, 2013 → 2025
BloombergNEF volume-weighted global average, nominal $/kWh. The cost of the EV's priciest part fell ~85% in a dozen years — and China sits below the global line.
2022's uptick reflects a spike in lithium and other metal prices; the decline resumed as those eased and LFP scaled. The gold marker is China's 2025 pack price ($84/kWh) — below the global average of $108. Source: BloombergNEF.20
Chapter Five · Product & Value

More Car per Dollar

BYD Seagull, a compact electric city car starting around $9,555
The BYD Seagull — roughly $9,555 to start, and the car that put advanced driver assistance within reach of an entry-level buyer.
No US or European automaker sells any new EV near this price.Photo: Ethan Llamas, CC BY-SA 4.0, via Wikimedia Commons

Dominance upstream becomes generosity downstream. In February 2025, BYD announced that "God's Eye," its highway-and-city driver-assistance suite, would roll out across its entire lineup — an initial 21 models spanning BYD, Denza, Fangcheng Bao and Yangwang — running on its in-house Xuanji A3 chipset alongside NVIDIA Orin and Horizon Robotics compute.6 Verified 2–1

The shock was the price. God's Eye C comes at no extra cost on the Seagull, a city EV starting at ¥69,800 — about $9,555. Before that announcement, the cheapest car in China with comparable L2+ assistance was the Leapmotor B10 at ¥109,800 (~$15,140), and BYD's own ADAS floor had been ¥216,000 (~$30,000). In one keynote, BYD cut the price of admission to advanced driver assistance by roughly two-thirds.6

Fig. 5 — Democratizing the Robot Chauffeur
Cheapest car with advanced (L2+) driver assistance in China
Starting MSRP at time of God's Eye launch, February 2025
BYD's prior ADAS floor2024
¥216,000  (~$30,000)
Leapmotor B10previous market floor
¥109,800  (~$15,140)
BYD Seagull + God's Eye CADAS included, no extra cost
¥69,800  (~$9,555)
Source: Counterpoint Research (Apr 2025), corroborated by Yicai, S&P Global AutoTech Insight, Fortune. Verified 2–1 (medium confidence — single primary analyst source). The ¥69,800 figure is the Feb 2025 MSRP; price-war discounting later pushed transaction prices lower still. BYD reaffirmed the full-line rollout in Dec 2025 and unveiled a next generation in May 2026.

There is no Western analogue. No US or European automaker sells any new EV near $9,500, let alone one with standard advanced driver assistance — the cheapest American EV transacts at roughly three times the Seagull's sticker. Meanwhile the battery beneath it, BYD's LFP Blade, is the same chemistry family that now powers a majority of the world's new EVs, including many Teslas — whose Shanghai-built cars ride on Chinese cells.1

0
models in God's Eye's initial rollout — across four brands, flagship to $9.5K city car
0
of EV batteries deployed globally in 2025 were LFP — the chemistry BYD's Blade pack made mainstream (<10% in 2020)
0
cheaper: the ADAS-equipped Seagull vs. BYD's own driver-assistance price floor a year earlier
Interactive — What Your Money Buys
More car per dollar — priced at home, where it's real
China-market prices, where Tesla and BYD compete head-to-head. In some classes the West has no entry at all. Tap through:
China manufacturer guide prices, base trim, pre-subsidy; converted at ¥6.77/$ (June 2026). Range is CLTC (more optimistic than EPA/WLTP). Mid-2026 prices are volatile amid China's price war — snapshots, not quotes. "No Western analogue" means no US/EU automaker offers a comparable model at any price near it; the nearest Western equivalents are noted. Sources: CarNewsChina, CnEVPost, Electrek, Edmunds, manufacturer sites. The earlier UK-price comparison understated the gap because export prices bake in shipping, duties and positioning — see the tariff note under each pairing.
Chapter Six · System Integration

The Machine That Builds the Machine

Why can BYD give away driver assistance on a $9,500 car and still out-earn Tesla? Because almost nothing in that car carries someone else's margin. BYD began life in 1995 as a battery maker; today it makes the cells, the cathode chemistry behind them, the power electronics, its own Xuanji ADAS chipset, the software stack — and it sells the surplus batteries to rivals. Each layer it owns is a markup it doesn't pay.

Thirty years, battery workshop to world's largest EV maker. Scroll the timeline:

1995
Wang Chuanfu founds BYD in Shenzhen as a rechargeable-battery maker, supplying mobile-phone cells.
2003
BYD buys a struggling state carmaker (Qinchuan Auto) and enters the car business — batteries first, cars second.
2008
Launches the F3DM, the world's first mass-produced plug-in hybrid. Warren Buffett's Berkshire Hathaway buys ~10% for $230M.
2020
Unveils the Blade Battery — a cobalt-free LFP pack that resets the industry's safety-and-cost baseline.
2022
Becomes the first major automaker to stop building pure combustion cars, going all-in on EVs and plug-in hybrids.
2023
World's largest NEV maker (~3M sold) and begins shipping cars on its own fleet of ro-ro car carriers.
2024
Out-earns Tesla on revenue ($107B) and overtakes it in quarterly battery-electric sales.
2025
"God's Eye" driver assistance goes line-wide; the Super e-Platform debuts; BYD seizes the global BEV crown and tops 1M exports.
2026
Targets 1.5–1.6M exports and launches in Canada — even as a brutal home-market price war squeezes profits.
Interactive — Anatomy of a BYD
What "makes everything itself" actually looks like
Tap a part to see which piece BYD builds in-house — and why that means no one else's margin rides along.
A schematic cutaway. Hotspots mark the layers BYD designs and manufactures itself — the basis of its cost advantage. Most Western automakers buy several of these (cells, chips, power electronics) from outside suppliers, each taking a margin.
BYD Auto manufacturing plant in Rayong, Thailand
BYD's plant in Rayong, Thailand — the model goes global: build the supply chain at home, then plant final assembly inside the tariff walls.
Overseas factories help BYD step around the import duties rising against it.Photo: iMoD Official, CC BY 3.0, via Wikimedia Commons

The integration shows up as speed. BYD spent RMB 63.4 billion ($9.2B) on R&D in 2025 — up 17% in a year when its profits fell 19% — and employs more than 120,000 engineers, with over 42,000 granted patents.10,11 When it decided every car should have city-capable driver assistance, the rollout took one product cycle, because the chips, cells, software and assembly lines all answer to the same company.

What we did not verify: several of BYD's most spectacular engineering claims — the 1,000-volt Super-e platform, megawatt "flash" charging, next-generation Blade batteries with 1,000 km range, cell-to-body construction, and its roll-on/roll-off car-carrier fleet — are widely reported but did not pass through this report's adversarial fact-check round. They are omitted from the figures above rather than presented as verified fact.
Chapter Seven · The Ecosystem

The Grid Beneath the Cars

A car is only as good as the place you can plug it in — and here China's lead is almost absurd. At the end of 2025 the country held more than 65% of all the public EV chargers on Earth — about 4.7 million of roughly 7 million — and built three-quarters of every new public charger added worldwide that year.21 Counting private points too, China passed 19.3 million charging points, up 52% in a single year.22 The cars didn't arrive and then wait for somewhere to charge; the grid was laid down to meet them.

And it isn't merely a big-country mirage. Adjust for population, or for the number of cars actually on the road, and China's charging network is still the densest of any major market — as the toggle below shows.

0
of the world's public EV chargers are in China — it built 75% of 2025's global additions
0
charging points in China as of Nov 2025 — up 52% year-on-year, ~3,560 public ones added a day
0
battery swaps performed by NIO alone, from 3,790 stations — about 3 minutes each
400km
of range in a 5-minute charge on BYD's 1,000 kW "megawatt" platform
Fig. 7a — The Charging Gap, In Scale
Public charging points: China vs. the West
Raw totals flatter big countries — so here it is normalized too. China leads on every measure. Toggle:
Bars rescale to the leader in each view. Denominators (end-2025): China 1.41B people, 366M cars, 43.97M NEVs on the road; Europe ~449M people, ~250M cars; US ~342M people, ~283M cars, ~7.5M EVs. A complementary measure — EVs per public charger — runs ~10 in China, ~13 in Europe and ~36 in the US (fewer = denser coverage). Sources: IEA GEVO 202621; NEA/EVCIPA22; China NEV fleet & 366M car parc26; US DOE AFDC (207,227 ports); ACEA/EAFO (~1.2M, EU). US EV stock approximate.

Then there is the speed. While the best Western fast chargers top out around 350 kW (Ionity, Electrify America) and Tesla's newest V4 hits 500 kW, BYD's 1,000-volt "Super e-Platform" charges at up to 1,000 kW — adding 400 km of range in five minutes, a rate BYD markets as "1 second, 2 kilometres." Huawei and Zeekr have gone further still, to 1.2–1.5 megawatts.25 China has set a national target of 100,000 ultra-fast public chargers by 2027.

Fig. 7b — Peak Charging Power
How fast the fastest chargers go (kW)
Top passenger-car charging power. Higher = more range per minute.
BYD Super e-PlatformChina · 1,000V, 400 km in 5 min
1,000 kW
Tesla Supercharger V4500 kW peak (Cybertruck); most ~325 kW
500 kW
Ionity / Electrify AmericaWestern public fast-charge ceiling
350 kW
BYD's car-side peak is ~2–3× the mainstream Western ceiling; its own chargers reach 1,360–1,500 kW, and Huawei/Zeekr offer 1.2–1.5 MW units. Caveat: BYD's flash-charging network is largely closed to non-BYD vehicles, and "400 km" is a CLTC marketing peak. Source: BYD via Electrek.25

And then there is the thing the West barely does at all: swapping the battery. Instead of waiting to charge, you drive into a station and a robot replaces your depleted pack with a full one — in about three minutes, faster than a gas fill-up. NIO has now performed over 100 million swaps from 3,790 stations, once hitting 158,290 in a single day.23

CHARGING BAY READY PACKS depleted fully charged A full battery in about 3 minutes

Swapping also rewrites the price tag. Under NIO's Battery-as-a-Service, you buy the car but lease the battery — cutting up to RMB 108,000 (~$16,000) off the sticker and turning range into a monthly subscription. And it is no longer one company's gamble: battery giant CATL built 1,020 "Choco-Swap" stations in a single year and teamed with oil major Sinopec to target up to 10,000 swap stations, with a national safety standard now in force.24 The state is treating refueling itself as shared infrastructure.

Interactive — The Refueling Race
How long to put 250 miles back in the tank?
A time-lapse: restoring ~400 km / 250 miles of range, four ways. Press play.
0:00
Elapsed · real time
Times to replenish ~250 mi (400 km) of driving range. Gas pickup ≈ 3 min (pump + handling). NIO battery swap ≈ 3 min for a full pack. BYD's 1,000 kW Super e-Platform: 400 km in ~5 min (CLTC marketing peak). Tesla V3 Supercharger (250 kW) ≈ 18 min for ~250 mi. Runs in real time by default — gas and swap finish at 3:00, BYD at 5:00, Tesla not until 18:00 — so you feel the gap. Use the speed toggle to fast-forward. The headline: China's battery swap and megawatt charging have pulled EV "refueling" to within striking distance of a gas pump, while conventional fast charging still takes 3–6× longer.
"The cars didn't arrive and then wait for somewhere to charge. The grid was laid down to meet them — 4.7 million plugs, megawatt chargers, and robots that swap your battery in three minutes."
The honest counterweight: scale isn't the same as efficiency. National public-charger utilization runs under 10% — and as low as ~2% in tier-1 cities — well below the level operators need to profit, so much of this network loses money. Rural coverage trails (about 80% of townships), paid rural charging only begins in 2026, and battery swap, for all its momentum, is still a niche: NIO's headline "80% of trips by swap" is a highway-only, single-brand, holiday-peak figure. Vehicle-to-grid remains mostly pilots (30 projects across 9 cities), not deployed capacity.22
Chapter Eight · The Counter-Case

What Could Still Go Wrong

A credible case states its weaknesses. The verified record contains real ones — all four below passed the same 3–0 fact-check as the dominance claims.1,13

i.

The price war is eating the home market

BYD's China NEV share fell from 34.1% (2024) to 27.2% (2025); its domestic retail sales dropped 6.3% and net profit fell 19%. The export boom is partly a release valve for a margin-crushing war at home.

ii.

Q1 2026: the crown wobbled

China's domestic NEV sales fell almost 25% year-over-year in Q1 2026 (~2.01M units); BYD dropped ~25% and Tesla retook the quarterly BEV lead. Annual leadership for 2026 is genuinely contested — even as Chinese EV exports more than doubled again.

iii.

Inventory is piling up abroad

CAAM-reported 2025 exports exceeded actual overseas sales by more than 25% — cars on ships and in lots, not in driveways. Export figures flatter the demand picture.

iv.

The tariff walls are rising

EU countervailing duties (in force to end-2029) held Chinese imports below 20% of EU EV sales. Mexico raised tariffs from 20% to 50% in January 2026. Southeast Asian duty waivers expired in December 2025. The US market remains effectively closed.

A Reader's Question, Fact-Checked

"How the numbers get padded" — and what the viral photos really show

A common challenge runs: I've seen aerial photos of airport-sized lots packed with thousands of unsold EVs — proof the sales figures are inflated. The instinct is right; the evidence usually cited is not. There are two different things here, and only one of them holds up.

● The real mechanism
"Zero-mileage used cars"

Automakers and dealers book a car as sold — registering and insuring it before any buyer exists — to hit targets and claim EV subsidies, then funnel the pristine, never-driven car into the used or export market. Reuters and CNBC documented Neta booking64,719 carsthis way over 15 months — more than half the 117,000 it reported. Zeekr was also implicated; Chinese state media called it out and Beijing moved to ban resale within six months of a "sale."

▲ The misattribution
The "EV graveyard" photos

The viral images of thousands of cars rotting in fields are mostly from 2019–2023 and show abandoned ride-hailing and car-sharing fleets from startups that collapsed when subsidies were cut — not unsold 2025 inventory. Bloomberg, Carscoops and The Drive traced them to dead fleets; France 24 debunked a batch falsely placed in Europe. Citing them as proof of current fake sales invites an easy rebuttal.

The honest version of the argument leans on the zero-mileage scandal plus the verified export-inventory overhang (counterpoint iii above) — not the old graveyard imagery. And note the limit of even the strong version: padding inflates domestic sales reports. It does not touch the supply-chain pillar — the >80% of cells, 90% of anode material and 93% of magnets are measured at the factory and refinery, not from sales filings.14,15,16

Fig. 8 — The Cost of Winning
Revenue quadrupled — then the price war bit profits
BYD revenue (bars) vs. net profit margin (line), 2021–2025. Volume soared; margin peaked in 2024, then fell as the price war cut deep.
Revenue rose from RMB 216B (2021) to RMB 804B (2025); net profit fell 19% in 2025 to RMB 32.3B as net margin slipped from 5.2% to ~4.0%. The squeeze deepened in Q1 2026 (net profit −55% YoY). Context: 227 models cut prices in China in 2024 (avg −8.3%), the average car price fell ~11% in two years, dealers lost ~RMB 178B in 2024, and in Feb 2026 regulators banned selling cars below cost. Sources: BYD filings via StockAnalysis/Macrotrends27; CnEVPost (price cuts, Q1 2026)28; price-war economics29.

Note what the counter-case does not touch. Tariffs, price wars and quarterly reversals contest the showroom. None of them moves the refinery layer: the 19-of-20 minerals, the 90% of anode material, the 93% of magnets.2,8 Quarterly leadership is volatile. The structural dominance, on the verified evidence, is not.

· · ·
The car race can still swing.
The supply chain already finished.
Chapter Nine · The Other Model

The Other Way: How the Nordics Did It

The West's most serious answer to China isn't in Detroit or Wolfsburg. It's in Oslo and Helsinki. The Nordics could never match China on scale — there are only about 27 million of them — so they did something else: they finished the transition first, and did it on clean power, by democratic consent. In 2025, 95.9% of Norway's new cars were fully electric, up from 54% in 2020 — the highest adoption rate on Earth, on a grid that is ~88% hydro.30

Fig. 9a — Adoption, 2020 → 2025
Who actually went electric
Share of new cars that are battery-electric. Color = how electric; circle size = annual volume. Toggle the year; tap a market.
0%100% BEV
BEV-only share of new-car sales (battery-electric, excluding plug-in hybrids), so figures are comparable. China's combined NEV share is ~54% (incl. plug-in hybrids); its BEV-only share is ~33% — but on a market 30× the Nordics'. Circle size reflects annual EV volume. The split this map makes visible: the Nordics are deep-red but tiny (highest share), China is huge but mid-red (biggest volume). Sources: OFV, ACEA, ICCT, CPCA, SMMT, IEA GEVO 2026.34

So is it the #2 EV market? By volume, no — not close. After China (~12.9M EVs), the #2 is Europe as a bloc (~2.6M battery-electric), and the #2 single country is the United States (~1.6M). All five Nordic countries combined sold about 431,000 BEVs — fewer than Germany alone (545,000), and roughly 3% of China's volume.33 But volume is the wrong yardstick for 27 million people. Normalize for size, and the Nordics aren't #2 — on adoption they're #1, and on the completeness of the transition, from clean power to near-total electrification, theirs is arguably the most impressive performance anywhere.

🇳🇴

Norway

The consumer end, solved

~96% of new cars electric, EVs now outnumber diesels on the road, ~88% hydro power, and >24,000 public chargers (~10,000 fast). Three decades of tax incentives, built by democratic consensus.

Caveat: the VAT exemption is being phased out from 2027 — the incentives that built it are now being dialed back.
🇫🇮

Finland

The supply end — the West's best shot

The closest thing the West has to China's upstream chain: Terrafame makes battery-grade nickel & cobalt sulphate (operating), Umicore runs the largest cobalt refinery outside China, and Keliber lithium + the Easpring cathode plant are being built — on a 95%-fossil-free grid.31

Caveat: Keliber's refinery is deferred on weak lithium prices, and the flagship cathode plant is 70% owned by China's Beijing Easpring.
🇸🇪

Sweden

The industrial bet — and its limits

Home to fossil-free "green steel" (HYBRIT, Stegra) for low-carbon car bodies, and to Volvo and Polestar.

Caveat: Northvolt — Europe's would-be battery champion — went bankrupt (the largest in modern Swedish history) and was sold to a US startup; Volvo (Geely-owned) saw 2025 operating profit nearly wiped out.32
🇩🇰

Denmark

The clean-power backbone

BEVs passed 60% of new sales in 2025, powered by the world's highest wind share — about 60% of Danish electricity. EVs charged on a grid that's already overwhelmingly renewable.

Caveat: like the others, a small market whose influence is as a model, not a manufacturing force.

And here is the twist that ties the whole story together: even the West's best answer runs partly on Chinese capital and technology. Finland's flagship cathode plant is majority-Chinese-owned; Volvo and Polestar are controlled by Geely; Sweden's home-grown battery champion collapsed. The Nordic model proves a clean, democratic EV economy is possible — but not yet that the West can build one without China somewhere in the chain.

"China built the biggest electric-car economy. The Nordics built the most complete one — for a thirtieth of the people, on the cleanest power, by vote."
In Sum · The Scorecard

The Verdict, Dimension by Dimension

Pulling the whole argument together: China and BYD lead on every front this report examined — but the kind of lead differs. Upstream, it is structural and, on the evidence, locked in. At the showroom, it is commanding yet genuinely contestable. Each card rates the lead and names the honest caveat.

Market dominance

Commanding

BYD is the world's #1 BEV seller (2.26M in 2025); China builds ~75% of all EVs.

But: Tesla retook the quarterly BEV lead in Q1 2026 — full-year 2026 is contested.

Supply chain

Commanding

>80% of cells, ~85% of cathode, >90% of anode material, 93% of rare-earth magnets.

But: Western capacity is being built — though the IEA still sees ~⅔ battery dominance in 2035.

Financial scale

Strong

Out-earns Tesla on both revenue (~$116B) and net profit ($4.7B).

But: profit fell 19% as the price war squeezed net margin to 4.1%.

Value per dollar

Commanding

The Seal undercuts the Model 3 ~28% at home; a ~$10k Seagull has no Western rival.

But: export markups and tariffs erase much of the edge outside China.

System integration

Commanding

Cells, cathode chemistry, chips, software — even the ships — built in-house.

But: top-tier ADAS compute still leans on NVIDIA / Horizon silicon.

Charging & infrastructure

Commanding

65% of the world's public chargers, megawatt charging, 3-minute battery swap.

But: charger utilization is under 10% — much of the network loses money.

Global reach

Strong & rising

1M+ exports; Chinese groups own Volvo, Lotus, Polestar, MG and more.

But: walled out of the US; EU, Canada and Mexico tariffs bite.

Overall

Structural lead

The upstream lead — minerals, batteries, the factory floor — is deep and durable. The downstream lead is real but contestable.

Bottom line: the car race can still swing; the supply chain already finished.
Coda · The Bigger Picture

Two Ways to Make a Future

Step back from the cars. The defining industry of the decade was won not by the freest market, but by the two most deliberately governed economies on Earth — sitting at opposite ends of the democratic spectrum. The same race, run on three different operating systems:

🇺🇸

United States

The market model

Light-touch and stop-start: the IRA's incentives switched on, then tariffs went up and EV credits were repealed. The state mostly deferred to the market.

The freest hand — the weakest result. A distant third, walled off behind tariffs.
🇨🇳

China

Authoritarian state capitalism

Decades of industrial policy, planning, subsidies and state-channeled credit — with ferocious private competition underneath. Built the supply chain and the cars.

Dominance — without democracy, and at the cost of a brutal price war.
🇳🇴

The Nordics

Democratic market socialism

High-tax, high-public-investment, union-dense, consensus-planned. Produced the world's fastest adoption and cleanest grids — by vote, not decree.

Dominance's cousin — achieved with more democracy, not less.

The lesson reads two opposite ways at once: deliberate state direction beat laissez-faire in the industry that mattered most this decade — and it did so whether the state was authoritarian or impeccably democratic. China and the Nordics are wildly different political economies, but they share the conviction the US has spent forty years arguing against: that governments should actively build the future, not merely clear the way for it.

The honest qualifications matter, and they cut both ways. This is a spectrum, not a binary — the US does industrial policy too (the IRA, the CHIPS Act); the contrast is degree and consistency. "Most successful" is a claim about translating state direction into industrial and social outcomes, not about every metric — the US still leads the innovation frontier and on aggregate wealth. And each winning model carries real costs: China's overcapacity, debt and the margin-destroying price war (and the absence of democracy); the Nordics' small scale, their reliance on global markets, Northvolt's collapse, and the fact that even their supply chain leans on Chinese capital. But the through-line holds. The country that bet the market would sort out the electric car came third. The two that planned for it came first — one without democracy, one with more of it.

The framing of China and the Nordics as two state-directed alternatives to the American market model draws on Nick Warino's The Nordic Model Invents the Goods (People's Policy Project, 2022), which argues that egalitarian, state-active economies are not less innovative but more.35