Europe's Green Paradox: Decarbonizing Into Dependence

The European Green Deal is the bloc's proudest claim to agency: here, at least, Europe leads, sets the standard the world copies, and builds the industries of the future. The claim runs aground on a distinction that is easy to state and hard to legislate around. Policy can move a demand curve almost instantly. A target, a mandate, a subsidy, and the demand for solar panels or battery cells or electrolysers exists by the end of the legislative session. A supply curve moves on a different clock entirely, the clock of permits, furnaces, refineries and trained chemists, measured in years and often a decade. Create the demand by decree and assume the supply will appear to meet it, and the supply does appear. It simply appears in the country that already built the factory. On current trajectory, every increment of European climate ambition is an increment of imports from China.

A demand curve at the speed of a vote, a supply curve at the speed of a furnace

The numbers describe a demand pull, not a domestic build. The European Commission and the European Court of Auditors put the bloc's reliance on China at roughly 98 percent for solar panels, around 98 percent for the rare-earth permanent magnets in every electric motor and wind turbine, 97 percent for magnesium, and close to 100 percent for processed rare earths. The figure that matters most is the one moving the wrong way: China supplied about 88 percent of EU electric-vehicle battery imports in 2025, up from roughly 75 percent in 2019. The mandate did its job and created the demand; the capacity to meet it at home did not exist, so the demand was met from abroad, and the dependence deepened in step with the ambition.

The reason the factory was never going to be in Europe is not mysterious, and it is not mainly about who holds the mine. China controls an estimated 90 percent of rare-earth and graphite refining and around 60 percent of lithium and cobalt processing, the midstream steps that are capital-hungry, energy-hungry and slow to permit. And European energy is dear: the Draghi report and the IEA put EU industrial electricity at two to three times US and Chinese levels and wholesale gas at roughly five times the US price through 2024. A deployment mandate does nothing to close that gap. It simply guarantees a buyer for whoever has already paid the fixed cost of the midstream, which is China. Diversifying the upstream, the dozen raw-material partnerships Brussels has signed, leaves the binding constraint untouched.

The two halves of the policy work against each other

Here is the part that turns an awkward outcome into a structural one. European green policy has two halves that pull in opposite directions. The deployment side, the renewables targets and the 2035 vehicle rules, raises the volume of clean technology that must be procured. The cost side, the carbon border levy and the due-diligence and reporting rules, raises the cost of producing that technology inside Europe. One half inflates the demand; the other handicaps the domestic supply. Run together, they push production out and pull imports in at the same time. A policy designed to build European industry is, in its own mechanics, an instruction to buy the output from somewhere cheaper, and the cheaper somewhere is the one the strategy was meant to escape.

The case histories are not anecdotes; they are the mechanism observed. BASF is cutting roughly 2,600 jobs at Ludwigshafen while completing an 8.7-billion-euro site in Zhanjiang, its largest investment ever. Northvolt, intended to supply about 13 percent of planned EU battery output by 2030, filed for bankruptcy in 2024 after raising more than 15 billion dollars. Meyer Burger, the symbol of European solar manufacturing, filed for insolvency in 2025, openly unable to compete with Chinese imports. ArcelorMittal shelved a 1.3-billion-euro green-steel conversion and returned the public subsidy. In each case the carbon, and the capability, did not disappear. It relocated to where the energy and the policy were friendlier.

The retreats are the revealed preference

When a legislated demand curve meets a capacity wall, a government has two choices: pay to build the capacity, or bend the target. Europe has consistently chosen to bend the target, and the choices of 2025 and 2026 are a clean record of the preference. The 2035 combustion-engine ban was diluted from a 100 percent carbon cut to 90 percent. The corporate due-diligence rules were deferred to 2029. The 2040 climate target now allows up to five percentage points to be met with carbon credits bought abroad. Most tellingly, the Critical Raw Materials Act, the instrument written specifically to build the missing capacity, was judged out of reach by the bloc's own auditors in February 2026: processing capacity near 24 percent against a 40 percent target, recycling at 1 to 5 percent against 25 percent, mines that take up to two decades to open. The Act named the destination, and the audit confirmed there is no route on the timetable.

The instruments meant to project power

That leaves the two tools meant to turn the policy outward into leverage, and both mostly turn inward into cost. The carbon border levy, in force from January 2026, taxes the European importer rather than the foreign producer and does not cover finished goods, so a Chinese vehicle, panel or battery crosses untouched while a European manufacturer pays for the embedded carbon in its imported inputs. The countervailing duties on Chinese electric vehicles, up to roughly 45 percent, were routed around before they bound: Chinese brands still doubled their EU share to about 6.1 percent in 2025 by pivoting to plug-in hybrids the duties do not cover, imports of which rose almost 900 percent, and by building inside the tariff wall, where a BYD plant in Hungary will produce duty-exempt cars from 2026. A lever whose terms the target sets is not leverage.

What would change this view

The argument is falsifiable, and it should be stated that way. We would revise it if EU midstream processing moved decisively toward the Critical Raw Materials Act trajectory rather than away from it; if a European cell, magnet or panel maker cleared commercial volume against Chinese imports without a permanent subsidy; if the industrial energy-cost gap with the United States and China narrowed materially; or if China's share of EU battery and clean-technology imports began to fall rather than rise. The striking thing is that the leading indicators have moved the other way: import shares up, champions bankrupt, targets relaxed. Until they turn, the mandate is a demand signal pointed at a foreign supplier.

What it reveals

None of this is an argument against decarbonisation, which remains necessary. It is an argument that Europe confused writing the rule with holding the capacity, and that the two have come apart in a way the policy's own structure guarantees. Read beside this series' note on Russian sanctions, the symmetry is exact and uncomfortable. Sanctions revealed the materials Europe could not afford to cut; green policy reveals the materials Europe cannot yet make. In both, the gap between the declared posture and the physical capacity is filled by someone else, and increasingly that someone is Beijing. Agency was never the ambition of the mandate. It is the capacity to meet it, and that is the line item Europe has declined to fund.