Test the Grand Continent. The ideas and analyses you won’t find anywhere else. But you’ll see them everywhere soon. In your inbox starting from eight euros per month
The December 11 will mark the anniversary of China’s accession to the World Trade Organization in 2001. You personally led the negotiations of the bilateral agreement between the Union and China in 2000. Twenty-five years later, has your view of that historic moment changed?
My analysis has changed, but not my initial choice.
The mistake was not China’s accession. The mistake was that the United States, the European Union and Japan did not strengthen the WTO rules on subsidies and state aid before Beijing’s accession, for the sake of preserving support for major national champions such as Boeing, Airbus or Toyota.
It is worth noting a little-known fact: at the end of 1999, two years before China’s accession, WTO members let the strictest provision of the subsidies agreement lapse, namely the presumption of serious harm when subsidies exceed 5% of production value.
While China paid a heavy price for joining the WTO by capping its average tariff at 10%, a level well below that of other emerging countries, the weakening of these WTO disciplines did not prevent the magnitude of Chinese public support from rising sharply since 2010.
But beyond these historical considerations, we now face a much more serious problem with China.
A much more serious problem?
The European Union’s trade deficit with China reached €360 billion in 2025, or nearly €1 billion per day.
The error was not China’s accession.
Pascal Lamy
Jens Eskelund, the president of the European Chamber of Commerce in China, describes the current economic relationship as “a giant 400-meter container ship loaded with 24,000 containers heading to Europe and returning almost empty.” This imbalance is accentuated as Chinese goods are partly diverted from the American market: in the first half of 2026, Chinese exports to the United States fell by 23% versus the first half of 2025, while exports to the EU rose by 5%.
This imbalance is specific to China. In 2025, the Union indeed posted a global goods trade surplus of €128 billion. Excluding China, Europe’s surplus is around €490 billion. Europe does not have a generalized problem of competitiveness in goods trade. It has a serious Chinese problem that it does not have with any other trading partner.
Is Europe’s China policy not changing?
It is evolving rapidly and, for many, if not most, the current trajectory is unsustainable.
But our way of addressing this problem remains structurally ill-suited.
Take a very recent example: on July 9, the European Commission opened an anti-dumping inquiry into Beijing’s duck meat imports.
The following day, China announced a temporary ban on all helium exports, a gas without which semiconductor manufacturing is impossible.
Peking justified this measure by disturbances in supply caused by the war in Iran, and that is plausible. But the difference in approach is highly revealing.
In what sense?
On the one hand, European trade defense operates product by product, under tightly framed legal procedures and under the scrutiny of the European Court of Justice, with a timetable that can stretch over many months, while Chinese administrative tools, discretionary, have an impact on an entire sector in a single day.
The EU’s trade deficit with China reached €360 billion in 2025, nearly €1 billion per day.
Pascal Lamy
The EU therefore faces, with tools designed for a completely different type of conflict, a rival that plans sector by sector and acts within a few days.
Things are starting to move in Germany, though…
Yes, and that’s new. The German Mechanical Engineering Industry Association (VDMA) now calls for countervailing duties by product groups and a reversal of the burden of proof. The head of Volkswagen, Oliver Blume, whose company led the charge against European countervailing duties in 2024, is now calling for duties on hybrid vehicles imported from China.
During the Franco-German ministerial council in July, Chancellor Merz said it was “self-evident that we must address this imbalance, because it comes at the expense of our industry,” and President Macron called for the Commission to act with emergency measures rather than lengthy investigations. The two governments committed to a joint plan on China, and preparatory discussions are already underway ahead of the visit of the European Commissioner for Trade, Maroš Šefčovič, to Beijing this month.
What does Beijing say in response to these criticisms?
Beijing essentially responds that this is simply a competitiveness gap and that talk of excess capacity is unnecessary.
Is that really the case?
Yes and no. It is true that China possesses legitimate comparative advantages thanks to the size of its market, a strategically determined industrial policy pursued with authority, and lower unit costs. But this assessment is incomplete.
The real question lies elsewhere. This economic giant has built its relationship with the rest of the world in a framework that creates macroeconomic imbalances that are structural.
In terms of GDP share, China’s surplus is not new: Germany’s current account surplus topped 7% of GDP for much of the 2010s.
However, China is the second-largest economy in the world, and a trade surplus of $1.2 trillion, as in 2025, has no precedent in absolute value.
Thus, this debate about comparative advantage tends to obscure three hard-to-contest realities.
Which ones?
First, subsidies, already noted. The scale of government support in China is unprecedented at this level.
According to OCED’s MAGIC database, between 2005 and 2024, Chinese-based firms received on average three to eight times more state support relative to their turnover than firms in OECD countries. Also according to the OECD, public support is responsible for about 60% of the growth in global market share of Chinese firms since 2005, and the IMF estimates that China’s public expenditure on industrial policy runs around 4% of GDP per year.
Europe does not have a generalized problem of competitiveness in goods trade. It has a serious Chinese problem.
Pascal Lamy
BYD, the car manufacturer, epitomizes real competitive advantages today. Yet, according to Kiel Institute, even BYD received direct subsidies totaling €2.1 billion in 2022, i.e. 3.5% of its turnover. Behind BYD are state-owned or state-supported manufacturers that enjoy even higher subsidies relative to turnover, whose survival does not depend on profitability. By August 2025, nearly 30% of Chinese industrial firms were loss-making, compared with roughly 20% before the pandemic.
More and more people are analyzing the structure of these imbalances through monetary policy. Do you share this conclusion?
Yes, partly, and that’s the second reality. Exchange rates signal undervaluation. China’s trade surplus should have pushed the renminbi higher. Yet the real effective exchange rate has depreciated by about 15% since March 2022.
Rigorous estimates of renminbi undervaluation range from 16% to 30%.
This operational mode is no secret: Chinese banks have resumed accumulating foreign assets to restrain the currency. And I share the views of Gita Gopinath, Pierre-Olivier Gourinchas, and Hélène Rey that there is no doubt the yuan is undervalued.
Whether this results from a deliberate choice to undervalue the currency or from the authorities’ control of domestic inflation, this undervaluation provides a substantial comparative advantage.
What is the third reality?
The weakness of domestic consumption. The long-promised rebalancing of production toward domestic demand has yet to occur.
As early as March 2007, Wen Jiabao formulated his famous “four no’s”: growth in China, he said, was “unstable, unbalanced, uncoordinated, and unsustainable.”
Since then, nothing has moved. Private consumption accounted for 39.8% of GDP in 2005, and it still accounted for 40.0% in 2024.
By comparison, in 2024 it accounted for 53% of GDP in the EU and 68% in the United States. The 15th five-year plan again addresses this issue, but simultaneously commits to unleashing “new-quality productive forces,” that is, to strengthening the investment- and export-led model.
The warning from Xi Jinping in 2021 about “avoiding the trap of ‘state assistance’ that nurtures laziness” perhaps best illustrates China’s situation compared to other official documents.
Do you think this is the result of a policy and a strategy?
Yes, but not only. The burst of the housing bubble wiped out household wealth to such a degree that Chinese families still suffer. Domestic car sales fell for ten months in a row up to July, while car exports surged by nearly 90%. And with total debt above 300% of GDP, Beijing seems to be pursuing deleveraging first.
How do you explain the Party’s policy?
Relying on consumption implies raising taxes to fund a welfare state capable of reducing precautionary saving or borrowing, and the Party rejects both options. Yet net exports remain the only growth source that does not add to debt.
That’s where Beijing’s argument fails on Ricardo’s own ground.
Is Ricardo really Xi Jinping’s guiding light? Or, in essence, with the doctrine of industrial maximalism, is the Party rather breaking the world economy…
Yes, China is in the process of killing the system of open global trade.
The Union faces, with tools designed for a type of conflict totally different, a rival who plans sector by sector and acts within days.
Pascal Lamy
In his Principles, Ricardo explained that a country that is better at everything sees that advantage absorbed by the exchange rate and wages, so that everyone ends up specializing in what they do best.
China has disrupted this mechanism. Capital controls and state intervention keep the currency undervalued, and a very high saving rate suppresses domestic demand. Growth in production that should have translated into wage increases and more imports has been redirected toward exports.
None of the key players in the international trading system anticipated this scenario?
No. Built in 1947 and reformed in 1995, it rested on a Ricardian logic. Each country should specialize, produce, and trade certain goods, and distortions would gradually adjust themselves, product by product, supplier by supplier.
The Asian model of the “flying geese” seemed to confirm this: Japan first specialized in the textile sector, then Korea and Taiwan took over before passing on their expertise to the rest of Southeast Asia. Comparative advantages shift, and countries abandon certain sectors as they climb up the value chain. No one imagined that a single country, simply by its size, would leave nothing for others.
Yet China continued to produce textiles and toys while developing its capacity in steel, chemicals, solar panels, batteries, cars, machine tools, semiconductors, and medicines. China’s share of global manufacturing value added rose from 8.5% in 2004 to 28% in 2024, while the EU’s share fell from 25% to 17%.
concretely, what does that mean for a European business?
That it no longer competes on equal terms with its Chinese rivals in a level playing field, whether in the Chinese market or elsewhere.
A European toolmaker is no longer merely competing with a Chinese peer using the same equipment; there is now a state with a long-term plan, a local administration determined to keep a company running to protect jobs, a public banking system willing to continue lending to support that company, whatever the commercial conditions or the cycle.
In Schumpeter, allowing “creative destruction” sends a useful signal on efficiency: the disappearance of a company signals that it has used its resources less efficiently than its competitors. But when a company survives primarily because it can secure public money, that signal fades, and the economic argument for competition free and fair loses its force.
Do you have an example?
Solar energy is the most striking example. In 2007, Europe produced almost 30% of the world’s photovoltaic cells. Today it produces only about 1%. It no longer produces a single silicon wafer, and only a few polysilicon plants remain. Kevin Rudd accurately described Beijing’s logic: these massive subsidies are temporary and will end once foreign competitors are eliminated; prices can then rise to the level that suits Chinese producers, since there will be no alternative supply sources. Restoring a normal situation could take a very long time, because it takes years, sometimes decades, to rebuild a supply chain, and in the meantime capital moves to safer havens.
Why isn’t traditional trade defense enough?
That’s an interesting question. One might think the issue is merely the speed of execution, but we do have a framework to address this extraordinary situation. That was, after all, the European Commission’s first instinct: to do the same thing, but more quickly.
In particular, no one had imagined that a country, by its sheer size, would leave nothing for others.
Pascal Lamy
It opened 33 trade defense investigations in 2024, a record since 2006, then 32 in 2025, while the historical average is around 12 per year. In a single year, the number of measures in force rose from 199 to 232.
The fundamental flaw is that the cost of processing rises as the number of cases grows. A system that requires new analyses, new injury investigations and new causality assessments for each product will always be disadvantaged against a disruption affecting an entire supply chain.
Additionally, the adjustments these instruments produce, roughly in the region of ten billion—at best—are out of proportion with the magnitude and speed of the deficit’s growth.
These tools also assume that firms seek to maximize profits. Is that assumption no longer valid?
That assumption does not apply to state-supported Chinese firms, which operate according to instrumental objectives and a strategy rather than shareholder demands. They can absorb unlimited duties to protect their market share rather than profitability, and that is precisely what they are doing today.
What should the Commission do in these conditions?
Act on volumes. In fact, it already has understood this: the 16 procedures opened against China in 2025 all rest on the finding of significant distortions in the Chinese economy, and that finding applies to a wide range of products.
More generally, what should Europe do? In a note just published by the Jacques Delors Institutes, you speak of Plan A and Plan B: negotiation, or the creation of a new framework.
Negotiations on trade and investment, which began on June 29, cover five areas: the rebalancing of trade and investment, export controls, intellectual property, WTO reform, and a joint mechanism to monitor trade flows. The first test occurred in September with hybrid cars: their numbers rose from 3,800 per month in October 2024 to 50,000 from July 2026, while they are subject to only a standard 10% duty. The European Commission asked Beijing to voluntarily limit its exports.
Europe is, I believe, at a turning point in its relationship with China: either Beijing agrees to rebalance the trade, or the Union will need to significantly strengthen its trade defense to protect its industry. We must give a real chance to negotiation. But Plan A is only valuable if backed by a credible Plan B, because China already seems to be betting that Europe will give way.
With Plan B already in preparation, what can we realistically expect from these negotiations?
Not much. The document outlining Beijing’s position offers, to date, no substantive concessions on the substance.
What Beijing can offer are purchases. China has signaled interest in additional equipment for semiconductor production, the single major European export likely to bring a bit of relief, but it is limited for security reasons. The likely outcome will be an insufficient and delayed offer, the implementation of which will take one or two years.
In 2007, Europe produced nearly 30% of the world’s photovoltaic cells. Today it produces only about 1%.
Pascal Lamy
In my view, three conditions must be met: verifiable limits on volumes exported in sectors where harm has been demonstrated; putting the exchange-rate issue on the agenda of the IMF and the G20; a commitment to production capacity in the sectors highlighted by Beijing in its anti-“involution” campaign.
It remains necessary to spell out the mechanism that will deliver a solution in the event Plan A fails, and to do so now.
On what institutional balance framework does your Plan B rest?
On a model that already exists: the steel regime, which came into force on July 1.
It was adopted in eight months only, a record for Brussels. Its stated objective is to address global overcapacity: the volume of duty-free imports was reduced by about 47%, and a 50% duty is imposed on everything above the quota. This is what Europe needs in all sectors exposed to similar structural shocks.
How would it work in practice?
The mechanism rests on a few principles. First, its justification: it tackles the detrimental effects of structural imbalances. What would trigger it? Investigations would be opened based on behaviors observed across an entire supply chain, and not line-by-line. These behaviors are known: a sustained rise in imports, prices below production costs, capacities that exceed domestic demand, high levels of public subsidies, and indicators of intervention to keep the currency low. Once the import volume is fixed, everything becomes automatic: once the quota is exhausted, the out-of-quota duty applies, with no new decision.
Why a tariff-rate quota rather than a simple quota?
Ultimately, it’s about quantities, but we use a tariff-rate quota: a low duty up to a certain volume, a higher duty beyond. The reason is legal, to stay within WTO rules; a pure quantitative restriction would violate Article XI of the GATT, while Article XXVIII allows renegotiation of consolidated duties, and thus the establishment of tariff-rate quotas. Volumes are allocated country-by-country based on historical flows, as during the 2018-2019 steel safeguard. This preserves market shares of those who played by the rules and imposes a known ceiling on Beijing.
Would you also shift from “rules of origin” to “rules of ownership”? What would that entail?
We would generalize the so-called “melt and pour” approach: a product manufactured with a Chinese input would still be considered Chinese, even if the factory is in Morocco or Hungary. Origin would be determined by ownership and composition, not by the final-origin label, and labeling rules would be tightened. This aligns with informal proposals from five member states: it is about preventing firms owned by Chinese capital, located in Morocco or Hungary and fed with Chinese inputs, from circumventing the regime.
Are these restrictions meant to last?
The anchoring in the multilateral system is permanent; European restrictions are not, even with this new structural balancing instrument. The regulation could schedule the phased lifting of quotas in two ways: a negotiated agreement on capacities or volumes to serve as the basis for quota allocation, or a confirmed return to balance.
You stress a political dimension at least as important as the economic: a solidarity fund to face retaliation.
Because retaliation will occur if Plan B is activated, and they will target precisely the Member States capable of weighing in a qualified majority.
A Plan A is valuable only if it is backed by a credible Plan B, because China seems already to be betting that Europe will yield.
Pascal Lamy
Before the Commission’s May debate, a Chinese media outlet published a targeted list of products: cosmetics, alcoholic beverages, meat and luxury goods. It concerns exports from France, Spain and Italy. This solidarity fund would be integrated into the same regulation and financed by revenues from duties on out-of-quota imports to compensate Member States and vulnerable sectors. It is said that the Commission is currently preparing such a mechanism. Indeed, if the costs are not distributed upfront, the threat underpinning Plan A is unlikely to be taken seriously in Beijing.
We must be aware of what this tool cannot do: it will not win the bottleneck battle. An outright prohibition of helium exports or a restriction on licenses for critical minerals could paralyze European production chains within weeks, and even a change in Europe’s trade rules would not alter that in the short term. This instrument is more limited and more feasible: it is designed to give European negotiators an additional lever by offering a fallback within the ongoing negotiations with China.
Five member states invoke national security to justify these protections. Why do you reject this approach?
Because the Union would betray its own principles by aligning with an American interpretation that invokes national security at every turn.
The European Union filed a complaint against the United States when Washington invoked national security to justify tariffs on steel and aluminum in 2018. It has consistently argued in Geneva that Article XXI is not solely a matter for each state’s judgment, and it has argued repeatedly that this article should be examined by the dispute settlement mechanism. Invoking this exception in response to overcapacity would place Europe on the side of the interpretation it has fought against for nearly a decade.
Which legal path should be preferred then?
The real objective is to renegotiate the 1995 tariff agreement, and GATT makes this explicit: Article XXVIII authorizes increasing consolidated duties if compensation is paid to the main suppliers. The Commission has invoked this for steel in October 2025, and the compensation took the form of country-specific contingents rather than a dispute.And what about currency? You mentioned this gap. Is it really possible to make it a trade policy issue?
It can be treated in exactly the same way, as a measurable behavior, not as a result. Gopinath, Gourinchas and Rey are right to say that turning the renminbi exchange rate into a condition of any agreement is a mistake, because the exchange rate reflects the entire Chinese economic policy, anchored in reserves, and “when you pressure for a reform of the exchange rate, you create a conflict.”
However, an intervention aimed at preventing currency appreciation is a quantifiable measure that can be the subject of a report, notably through state banks. China continued to accumulate foreign reserves in 2025. So this can trigger the system, with the real exchange rate as the criterion for lifting measures in the next phase.
Where to start?
The European Council should first exercise the powers conferred on it by Article 219 of the Treaty on the Functioning of the European Union — which has never been used — to develop general guidelines on exchange-rate policy toward third countries. Europe should then inform the IMF and the G20 of this position. Europe cannot ask others to debate the monetary question if it is not prepared to do so itself.
The final element of your proposal, and perhaps the most striking, is the advance-defined exit condition. Why is this so important?
A reasonable benchmark would be this: China’s industrial surplus returning to around one-tenth of global merchandise imports, i.e., the level it reached before the housing market collapse.
Exit criteria require verifiable changes in data, and not merely a series of statements of intent. Publishing them in advance is what a rule-abiding player would do when wielding power conferred by those same rules.
It explains to its partners what it does, why, and when it will stop.
The steel regime provides a revision date; the instrument proposed here, by contrast, would have a numerical exit target.
Twenty-five years on, what lessons can we draw for the future of the system itself?
Europe is currently the world’s largest single market with strong purchasing power and still open to Chinese products. Beijing therefore needs this access, and that is why it responds with threats rather than proposals. Europe is also the largest economy still attached to a rules-based trading system, which is why other countries such as Japan, South Korea or Canada watch closely how it handles this crisis and why the approach is as important as the objective.
The architects of the 1995 agreement were convinced that adjustment would happen by itself: if a country overproduced, prices, wages and the value of its currency would rise until it reduced production. That mechanism does not work for the world’s leading manufacturing power because the rules were not written for a member capable of neutralizing them.
Europe must use the instruments that already exist within the system with prudence and transparency, and develop missing regulations with those who are willing to accept them, rather than waiting for the situation to evolve or for others to formulate new rules. It must also improve its competitiveness, starting with the swift completion of the internal market, where holes remain in finance, energy and connectivity. We are working on this with Enrico Letta at the Jacques Delors Foundation Friends of Europe and our campaign “One Europe, One Market for All.”