The idea of Canada joining the EU has been discussed with unusual seriousness in recent months. France’s foreign minister, Jean-Noël Barrot, raised it in March. Finland’s president, Alexander Stubb, even told Mark Carney personally that he should consider it. In an April YouGov poll, 55 per cent of respondents in Germany and 51 per cent in Spain said Canada should join the EU. In Poland, the figure was 41 per cent. A Nanos poll found that 57 per cent of Canadians supported the idea. Carney’s position, however, has been clear: closer ties with the EU, yes; membership for now, no.

The obstacle is not political. It is legal. Article 49 of the EU Treaty says that only European states can become members of the EU. Changing that rule would require the unanimous agreement of all 27 member states, followed by ratification in each of them. Only then could Canada apply under the rules of Article 49 itself: a unanimous decision by the Council, the consent of the European Parliament and ratification of the accession treaty by all member states.

A new member of Canada’s size, and the prospect of an enlarged single market, may sound attractive. But few member states would embark on the laborious process of changing the treaties without knowing what the outcome would be. The polls therefore measure enthusiasm for an idea that is unlikely to become a political question any time soon.

***

Canada has already moved further in two other areas.

In February 2026, it signed the SAFE agreement at the Munich Security Conference. SAFE is an EU financing instrument worth €150 billion for joint defence procurement. Canada became the first non-European country whose companies can participate in certain joint procurement projects financed through SAFE.

Then, in May, Mark Carney attended the eighth summit of the European Political Community (EPC) in Yerevan. He was the first head of government from a non-European country to attend an EPC summit. There he described Canada as “the most European of non-European countries” – a line made for quotation and, at the same time, carefully non-committal. The Canadian government does not describe either SAFE or participation in the EPC as a step towards EU membership. Nor does it need to: neither involves a vote on membership.

The EPC emerged in 2022 from a problem the normal accession process could not solve. How could countries be brought closer to Europe without making them wait years for membership?

Ukraine had applied to join the EU just days after Russia’s full-scale invasion. The application could not offer any quick route to accession. Macron therefore proposed a different format, one that could bring Kyiv, the United Kingdom, which had already left the EU, and the Balkan accession candidates closer to Europe without opening their accession procedures.

The first meeting took place in Prague in October 2022. The EU-27, Ukraine and the UK sat together at the table; 44 European countries took part altogether. The new platform had neither a legal basis in the EU treaties nor a fixed membership.

In September 2022, Ursula von der Leyen had described the initiative as a way of engaging European countries beyond the normal accession process. The institutional framework was left open.

***

That open-ended design has now acquired a new use.

When Macron presented the EPC in 2022, it was explicitly framed around the European continent. The forum was meant to bring together Ukraine’s ambitions and Britain’s estrangement within a deliberately undefined European space.

Now Canada, a G7 country from outside Europe, is sitting at the table for the first time. No vote was needed to extend the forum to non-European countries. All it took was an invitation and an acceptance.

That may sound like a small institutional detail. It is not. The EU does not have to change its treaties to bring Canada closer politically. It does not have to decide whether Canada is “European enough” for membership. It can simply use a political space whose boundaries were never fully defined.

There is still a hard limit. In 2024, 75.9 per cent of Canada’s goods exports went to the United States; in 2025, the figure was 71.7 per cent. Geography is what it is. No gesture towards Brussels will change Canada’s existing supply chains in the short term.

Something else is changing, though. It is a question of trust. With whom does Ottawa want to sit at the table when decisions are made about defence procurement and military security? Canada remains economically and militarily closely tied to the United States. At the same time, it is beginning to build a second strategic relationship in Europe.

The two directions are not necessarily contradictory. Canada can remain dependent on the United States for trade while seeking greater strategic room in Europe. Brussels, for its part, can widen its political reach without spending the political capital required to change the treaties. Ottawa can demonstrate that it is diversifying its strategic relationships without pretending that its economy could quickly detach itself from the United States.

THE VERDICT

No one is going to change Article 49. Persuading 27 governments to redefine the term “European state” and secure the necessary national ratifications is not a battle Brussels will take on unless a crisis forces it to. Unlike in the case of Ukraine, there are no tanks on Canada’s borders.

The more interesting decision is being made elsewhere: in a political forum with no legal basis for EU membership, created four years ago in response to a different crisis. Its open structure now permits something that was never explicitly ruled out when the forum was created.

Canada is unlikely to join the European Union any time soon. But it has begun to take a seat at a European table where proximity is possible without membership.

That may be the more realistic form of European enlargement in the 21st century: not enlarging the EU, but enlarging the political space that Europe can encompass.

Editor’s note on AI: I use AI tools for translation, English-language editing and, where useful, structural feedback. The arguments, analysis, sources and conclusions are my own, and I take full responsibility for the published text.

I have disabled Substack's AI detection because I have found that it produces widely varying scores for my writing that do not meaningfully or fairly reflect how these articles are written or the work behind them. I share the concerns of many other Substack authors and would rather write for my readers than for an AI detector.

Eight years later, Germany, Italy, and Austria were all chartering floating LNG terminals of their own, in a hurry, after the pipeline gas their industrial models had depended on stopped flowing. The insurance Lithuania had been mocked for buying in 2014 turned out to be the policy half the continent needed in 2022.

· · ·

The easy conclusion is that the Baltic states saw the future and Berlin did not. It is also the wrong conclusion. Estonia, Latvia, and Lithuania did not have better information about Russia. What they had was a different relationship to being wrong. For Germany, a misjudgment about Russian reliability was an energy pricing problem. For the Baltic states, the same misjudgment is existential. When the cost of error is survival, you act on warnings that larger countries can afford to file away.

A large state believes it can absorb almost any shock, so it treats hedging as an unnecessary expense. A small, exposed state knows it cannot absorb the shock, so it pays the premium early, when the premium still looks absurd. The information available to both is identical. The behavior is not, because the price of being wrong is not.

Smallness alone does not produce this. Denmark is small. Portugal is small. Neither read Russia earlier than anyone else, because neither sits where the Baltic states sit. It is the combination, small and exposed, that removes the luxury of optimism. The Baltic states are not clairvoyant. They are simply the part of Europe where wishful thinking about Moscow was never affordable.

· · ·

The pattern extends well beyond energy. On February 9, 2025, the three countries disconnected their electricity grids from the Russian-controlled network they had been wired into since the Soviet era and synchronized with continental Europe instead. The switch took more than a decade of preparation and considerable money, and it was carried out while much of Europe still described interdependence with difficult neighbors as a source of stability. The Baltic view was older and colder: a grid your adversary controls is a switch in someone else’s hand.

The sharpest test, though, came from China rather than Russia. In 2021, Lithuania allowed Taiwan to open a representative office in Vilnius under the name “Taiwanese” rather than the diplomatically safer “Taipei.” Beijing responded with the heaviest economic pressure it has directed at any European state: Lithuanian goods stopped clearing Chinese customs, and multinational companies reported pressure to remove Lithuanian components from their supply chains.

Except that the pressure failed. Lithuania survived it, essentially intact, because China accounted for less than one percent of Lithuanian exports. There was very little to coerce, and that was not luck. A country that knows it cannot afford dependency does not build it in the first place, and a country without dependency turns out to be remarkably difficult to blackmail. Compare that with Germany, France, and the Netherlands, which spent the same decades deliberately deepening their exposure to the Chinese market, each on the theory that a large, diversified economy could absorb a risk a small one could not. Volkswagen no longer has a China policy. It has a China exposure, and the exposure sets the policy. ASML and the French luxury conglomerates made the same bet in different sectors, for the same reason: the market was too large to walk away from, and the downside looked too diffuse to price.

· · ·

Lithuania has now lived both halves of the question every diversifying economy eventually asks: what hedging costs before the crisis, and what it is worth during one. The cost was real, a terminal bought before it was needed, a decade of grid investment, a bruising and initially pointless-looking confrontation with Beijing. The value was also real. When the test came, Lithuania was the one country in the exchange that could not be squeezed.

None of this makes small states models to copy. Their caution was never chosen; it came with the geography, and countries with more weight and more distance will always be tempted to believe that size is itself a form of protection. Sometimes it is. But the past decade suggests a different way to read Europe’s map. What looked from Berlin like Baltic paranoia in 2014 was something more ordinary: three small countries pricing a risk the rest of the continent had decided not to see. The risk did not care who priced it.

THE VERDICT

The logic does not stop at Vilnius. Strategic autonomy remains, in Brussels, a doctrine no government has funded for the same reason Berlin bet on Russian gas and Beijing’s demand: the cost of hedging is concentrated and immediate, the benefit is diffuse and arrives after the next election. Large economies can defer that trade almost indefinitely, because deferral rarely costs anyone a seat. Small, exposed states cannot defer it, which is why the Baltics built a terminal nobody needed and left a grid nobody was forcing them to leave. The choice was never about foresight. It was about who gets to wait, and for how long. Somewhere between Taiwan’s chip fabs and Europe’s own unbuilt defense-industrial base, the same wager is being made again, by capitals that have not yet had to find out what their premium would have cost.

This piece is adapted from "The Luxury of Being Wrong," published in Rajasthan Patrika on 27 July 2026 as part of the monthly "Letter from Europe" column. Rajasthan Patrika, one of India's largest Hindi-language daily newspapers, carries the column on a monthly basis.

· · ·

Article 2 is not an Israeli specialty. Since the mid-1990s, the EU has written near-identical clauses into virtually every trade and association agreement it signs, from the Cotonou framework with 79 African, Caribbean and Pacific states to the accords of its neighborhood policy. The logic was contractual: Europe grants market access, the partner accepts human rights as an essential element of the deal, a violation gives Europe the right to suspend. No other power writes its values into its commercial treaties this systematically. It is the closest thing the EU has to a signature.

The Israel case should have been the clause’s easiest test. The EU is Israel’s largest trading partner, taking 31.7 percent of its total goods trade in 2025, worth €43.3 billion, and its largest foreign investor, with member states holding €72.1 billion in direct investment against €39.2 billion for the United States. The dependence runs one way: Israel ranks 27th among the EU’s trading partners, at 0.8 percent of the bloc’s trade. Suspending preferences would have been painful for Israel and nearly invisible in Europe. If economic leverage were what the clause runs on, no partner offers more of it.

The clause does not run on leverage. Under the Cotonou Agreement, the essential-elements procedure has been applied roughly fifteen times since 2000, against Fiji, Zimbabwe, the Central African Republic, Guinea-Bissau, Togo, Madagascar, Burundi. Aid was cut, cooperation frozen, governments pressured. The clause works, repeatedly, against partners that are small, poor, and without an advocate in the European Council. It has never been enforced against a partner whose sanctioning would cost a large member state something at home.

· · ·

And how little was asked. Von der Leyen’s proposal did not suspend the agreement. It would have reintroduced tariffs on 37 percent of Israeli exports to the EU, falling mostly on agricultural goods, dates, fruit, nuts. Too small to move an economy built on technology and chemicals; the Commission, anticipating resistance, had calibrated the penalty down to a gesture. The gesture still found no majority.

Among the governments blocking it, Germany is the one that states its reasoning openly: the historical responsibility Berlin calls Staatsräson, translated into dialogue without consequence. “We continue to rely on direct dialogue with Israel,” Germany’s minister of state for Europe said in May 2026, eight months after the Commission had put its finding of breach on the table. Italy’s position is softer in form and identical in effect: Rome will consider trade measures only if they do not harm the Israeli civilian population, a condition that few trade measures could ever meet, which may be the point.

Then Hungary’s government fell, and the mechanism showed itself completely. Viktor Orbán had been the loudest of the blockers. His departure was supposed to unlock European action, and in May 2026, action came: sanctions against violent settlers in the West Bank. The two ministers on the Commission’s original list, Bezalel Smotrich and Itamar Ben-Gvir, were no longer on it. The blocker fell, and the proposal shrank to meet the next-most-reluctant government. European foreign policy does not follow its majorities. It follows the last hesitant capital. Remove one veto and the system does not act; it adjusts downward.

The standard defense is that the clause still works as a threat, that the September proposal helped push Israel toward the October 2025 ceasefire. The record reads differently. Within days of the ceasefire, the Commission’s own spokespeople were suggesting the proposals might be withdrawn if “the context changed.” A threat that begins negotiating against itself at the first sign of movement is not held in reserve. It is being walked back, and everyone watching, in Jerusalem and everywhere else, can see the direction.

· · ·

Everywhere else is where the cost lands. The EU is negotiating or ratifying agreements with Mercosur, India, Indonesia, and the Gulf states, and each text will contain the same essential-elements clause, because the clause is non-negotiable in form. Every one of those governments has now watched a formal finding of violation produce a shrunken penalty and then no decision at all. They have learned what the clause actually says: enforcement depends not on the violation but on the violator’s friends.

THE VERDICT

Brussels describes this as a failure to act on Israel. It is larger. A conditionality that applies only to the weak is not conditionality with gaps; it is a hierarchy with paperwork, and the EU has now certified its own hierarchy in public, a documented breach on one side, fifteen enforcement cases on the other. The clause went untested for thirty years because no violator was ever strong enough to test it. Now one was, and the answer is on file for every future signatory. Fiji, Togo, Guinea-Bissau, Madagascar: the list of countries against which Europe’s principles have actually been enforced is a list of countries that could not do anything about it. That is not where the EU claims its values live. It is where they have a record.

The scene is usually read as a verdict. Europe has fallen behind, and even Germany knows it. That reading is not wrong, but it answers a question that has been put badly. Everyone asks whether Europe has lost the artificial intelligence race, and measured one way, the answer is plain. The European Union’s flagship plan aims to mobilize some 200 billion euros for artificial intelligence by the end of the decade. Five American companies expect to spend around 700 billion dollars on AI infrastructure in 2026 alone. No roadmap closes a gap of that size. If the race is to build the most powerful models and the chips beneath them, Europe is not behind so much as in a different category.

But that is one race, and it is not self-evident that it matters most.

···

The economic value of a technology rarely comes from inventing it. It comes from spreading it through the economy, from the unglamorous business of putting it to work in factories, hospitals, offices, and supply chains. A model that no one uses changes nothing, while an ordinary tool deployed everywhere can reshape an economy. The prize in artificial intelligence will go disproportionately to whoever applies it widely, rather than to whoever announces it first.

On that measure, Europe’s position looks different and considerably less hopeless. Western European factories use more industrial robots per worker than North American ones, and Germany is the third most automated economy on earth, behind only South Korea and Singapore. This is not a continent that fears machines that replace human labor. It has spent four decades installing them, for the most part by negotiating their arrival with its workforce rather than imposing it. The picture of a timid, technophobic Europe does not survive contact with the data.

The same shift is reaching the office. The share of European companies using artificial intelligence climbed from 13.5 percent in 2024 to 20 percent a year later, and in Denmark, Finland, and Sweden, it now runs above a third. Most of that is the ordinary cognitive work of a service economy: sorting text, handling customers, drafting, and searching. Where Europe leads, it does so on the strength of trained workforces, dense digital infrastructure, and a short distance between a firm and the skills it needs, not on lighter regulation.

Why this matters more for Europe than for anyone is demographic. A shrinking workforce can sustain its pensions and living standards only by producing more from each remaining worker, which is what applied artificial intelligence delivers. Europe needs that productivity more than the younger United States, and far more than India, which has workers to spare and the opposite problem. The continent with the weakest demographics has the strongest reason to be the world’s most aggressive adopter.

···

So Europe’s problem is one of attention. It is pouring scarce political energy into the race it cannot win while neglecting the one it might. In June, the European Commission unveiled a sweeping package to build sovereign European clouds, chips, and models, and to reserve sensitive public contracts for European suppliers. The instinct behind it is to catch up in inventing, to wall off a market and raise champions inside it. Almost none of it addresses the duller and decisive task of closing the gap between firms that already use these tools and those that do not, or between large companies and small.

The competition there is real. China is aging too and has made deployment a national project, and 87 percent of the humanoid robots delivered last year were built there. Japan, older still, turned to automation early and deliberately, in large part to offset a workforce it knew would shrink. The track Europe could plausibly run is not empty, and it is the one whose prize Europe actually needs.

Merz was right to be impressed in Hangzhou. The machines were genuinely remarkable, and they were Chinese. But the factories most ready to put such machines to work are still in Germany, and the economies that most need what those machines produce are still in Europe. The danger is not that Europe cannot compete, but that a continent which needs the productivity more than anyone has decided to prove it can build the machines instead of using them.

THE VERDICT

Europe is making a category error it can still correct and a sequencing error it might not. Two hundred billion euros aimed at chips and models will not close a gap measured in the hundreds of billions, and chasing it mainly advertises the chase. Diffusion offers no equivalent spectacle. There is no flagship to fund, no national champion to photograph, no single contract that proves a government is serious about it. Adoption happens unevenly, firm by firm, mostly invisible to anyone outside the firm itself, and that invisibility is exactly why Brussels keeps choosing the other race. A political system built to reward leaders who can point at something prefers the contest that produces a building over the one that produces a percentage point of productivity nobody can put a ribbon on. The continent with the most urgent demographic case for adoption is behaving like one that still has time to invent its way out, and it can keep affording that mistake for exactly as long as nobody asks what the 200 billion was actually supposed to fix.

This piece is adapted from “The Wrong Race,” published in Rajasthan Patrika on 16 June 2026 as part of the monthly “Letter from Europe” column. Rajasthan Patrika, one of India’s largest Hindi-language daily newspapers, carries the column on a monthly basis.

What followed was not decline, it was something harder to see and harder to reverse: the quiet institutionalization of low expectations.

· · ·

The standard explanation for Europe’s sluggish growth points to regulation, fragmented capital markets, an underdeveloped venture capital sector. All of this is real. But it mistakes the symptom for the cause. Europe did not end up with these structures by accident. It chose them, repeatedly, through the normal operation of democratic politics. Understanding why requires looking not at Brussels, but at the ballot box.

Europe’s electorates are among the oldest of any major economic bloc. The details vary — Italy older, France somewhat younger, the Nordic states somewhere between — but the direction is uniform. Across the continent, the median voter is older than a generation ago, will be older still a generation from now, and is making perfectly rational choices that, in aggregate, produce something that looks from the outside like collective irrationality.

Older electorates have more to protect and less time to benefit from disruption. They prefer transfers over investment, continuity over competition, the pension guarantee over the startup ecosystem. They do not vote for managed decline. They vote for stability, and managed decline is what stability produces when the underlying demographics no longer support the model.

· · ·

The evidence is not in the headlines but in the budget lines. Germany transfers more than €100 billion annually to its pension system — roughly a quarter of the entire federal budget. That figure has grown every year for a decade, essentially not because the system became more generous, but because the electorate it serves became larger. More retirees, more votes, more transfers. The arithmetic is not complicated and neither is the politics.

This is the mechanism in its simplest form: the voter who benefits from a transfer is present, votes consistently, and knows exactly what is at stake. The taxpayer who will fund it in twenty years is either young enough to be indifferent or not yet born. Democratic systems are not designed to resolve that asymmetry. They are designed to reflect it.

Across Europe, the same trade-off resolves the same way. In France, pension reform provoked mass protests that forced a government to bypass parliament to pass legislation it had been elected to deliver. In Italy, national debt approaches 140 percent of GDP while successive governments have found no majority for the measures that would reduce it. The instruments differ but the outcome is consistent: present costs are deferred, future burdens grow, and each decision is legitimate in isolation while the accumulation belongs to no one.

· · ·

The growth consequences compound quietly. An economy expanding at one percent rather than two loses more than output. It loses the margin that makes reform tolerable. When growth is strong, restructuring has a cushion: those who gain can compensate those who lose. At one percent, every reform is a zero-sum argument about who bears the cost today. The political system responds predictably. It avoids the argument.

The trap is self-reinforcing: low growth makes reform harder, the difficulty of reform keeps growth low, and the aging electorate votes to maintain the existing distribution for as long as possible — which is to say, for as long as they are alive to vote.

· · ·

None of this means Europe is finished. It means Europe has arrived at a condition it has not yet been forced to acknowledge honestly: that its political systems are optimized for the preferences of voters who are here, at the expense of an economy that needs to work for people who are not yet old enough to vote, or not yet born.

THE VERDICT

Europe’s growth problem is not a governance failure. It is democracy working as designed — producing exactly the outcome its incentive structure predicts. The aging voter who prefers transfers to investment is not irrational. The politician who delivers them is not negligent. What Europe has not done is name what this produces: a compound deficit, accumulating across decades, that each generation inherits slightly worse and slightly less able to reverse. At some point, the invoice arrives. The people who will pay it are already in school.

This piece is adapted from "The Continent That Forgot to Grow," published in Rajasthan Patrika on 19 May 2026 as part of the monthly "Letter from Europe" column. Rajasthan Patrika, one of India's largest Hindi-language daily newspapers, carries the column on a monthly basis.

Arm is not a cautionary tale. It is the operating model.

· · ·

The standard diagnosis of Europe’s technology problem runs as follows: the continent produces talented engineers and promising startups, but cannot scale them into global companies because its markets are fragmented, its regulation is hostile, and its risk culture too conservative. Draghi said it. The European Commission has been saying it in various formulations since the dot-com era.

The diagnosis identifies a real constraint. It mislocates the primary failure.

Europe’s problem is not that it cannot build technology companies. It is that at the moment those companies become genuinely valuable, ownership transfers — to American funds, American acquirers, or American exchanges — and the control rights over future investment, expansion, and strategy transfer with it. A European researcher develops a technology. A European startup commercializes it. An American fund provides the growth capital, on terms that embed American exit expectations. An American exchange or acquirer provides the liquidity event. The equity, meaning the claim on future earnings and the right to direct how they are deployed, ends up in California or New York.

Europe retains the wages. It transfers the compounding.

· · ·

This exit structure is not a founder preference or a cultural failure. It reflects a capital allocation system that makes it structurally rational for European long-term investors to finance American markets rather than European growth assets.

Under Solvency II, the EU’s regulatory framework for insurers, unlisted private equity carries a standard capital charge of 49%, compared to 39% for listed equities in deep, liquid markets. At that differential, holding growth-stage stakes in European technology companies requires nearly half the invested amount to be set aside as regulatory capital. European insurers allocate a fraction of a percent of their total assets to private equity — a share that has remained near zero for a decade despite repeated policy efforts to change it, and roughly half the share that pension funds deploy, despite comparable liability horizons. The institutional capital that should be the natural long-term shareholder of European technology companies is, by regulatory construction, pointed elsewhere.

Where it goes is not hard to trace. American equity markets are deeper, more liquid, and carry no equivalent capital penalty for institutional holders. A European insurer optimizing its Solvency II position rationally overweights US listed equities. European long-term savings flow into American capital markets, which return to Europe as acquisition finance — on terms that transfer ownership outward. Europe provides the savings and the early-stage risk. The control rights move in one direction.

· · ·

ASML, SAP, and Adyen demonstrate that European capital can retain technology companies at scale. What they share is that each operates in a sector where technological irreplaceability or deeply embedded customer relationships make foreign acquisition unattractive. ASML makes lithography machines no one else can manufacture. Its ownership is partly European not because the capital markets functioned correctly, but because the product provides a structural exemption from the normal exit logic. That exemption is not available to most European technology companies, and it cannot be manufactured by policy.

The Capital Markets Union has been official EU policy since 2015. Eleven years of roadmaps have produced incremental harmonization of prospectus rules and marginal improvements in cross-border fund distribution. The gap persists because closing it requires member states to surrender national financial preferences — their domestic exchanges, their banking champions, their regulatory autonomy — that none has been willing to part with. France does not want Frankfurt setting the terms. Germany does not want Paris. The result is 27 partial markets instead of one deep one, and a venture capital ecosystem that remains structurally underweight in the late-stage categories where ownership is actually determined.

The Commission’s preferred response is more innovation policy: larger grants, deeper public funding, more startup support programs. Producing more startups for foreign capital to acquire at scale is not an industrial strategy. It is a subsidy program for foreign acquirers.

· · ·

THE VERDICT

Europe does not have a technology problem. It has an ownership problem, and the two require entirely different solutions. The Solvency II capital charge on unlisted equity is not a conspiracy against European technology. It encodes a political choice — policyholder protection over productive investment — that member states have made and could unmake. The Capital Markets Union has been a roadmap for eleven years because what it actually requires is that France, Germany, and Poland give up the national financial architecture each has spent decades protecting. More innovation funding is the answer that avoids that question. Europe has been choosing it since 2015, and the equity keeps leaving.

Subscribe now

Italy’s fiscal watchdog puts the debt-to-GDP ratio at 137.1 percent for 2025, rising to 138.6 percent in 2026. According to IMF estimates, Italy’s ratio is set to reach 138.4 percent this year, compared to 136.9 percent for Greece. The watchdog ran thousands of statistical simulations of those projections. Roughly half produced worse outcomes than the government forecast. The planned privatization revenues — nearly €20 billion over three years — were flagged as optimistic by the same body that noted Italy has consistently missed earlier privatization targets. The independent court of auditors has previously described such plans as window dressing. Eunews

This is not a story about Italian mismanagement. It is a story about what happens when a monetary union is built without a fiscal one, and what institution ends up holding the consequences.

· · ·

On September 8, 2022, the European Central Bank raised its key interest rate by 75 basis points, the largest single move in its history to that point. In Frankfurt, economists praised the decisive shift. In Rome, a different calculation had already begun.

Italy’s public debt now stands at over €3 trillion, 137 percent of GDP. To illustrate what interest rates mean at this scale: in 2021, when the average rate on Italian debt was around 1.5 percent, the annual interest bill was manageable. At four percent — a level that prevailed in Europe before the financial crisis — Italy would pay roughly €120 billion annually in interest alone, more than its entire education and defense budgets combined. The country issued over €550 billion in new and refinanced debt in 2025 alone. Each percentage point of rate increase translates immediately into billions of additional servicing costs, with no offsetting policy lever available. Pravda EUMEF

Within 24 hours of the ECB’s September 2022 announcement, the spread between Italian and German government bonds had widened sharply. Markets were testing how far Frankfurt would go. The answer had already been given weeks earlier: on July 21, Christine Lagarde had announced the Transmission Protection Instrument, a program for unlimited bond purchases should spreads rise in ways deemed “unjustified.” The announcement alone calmed markets immediately.

The signal was unambiguous. The ECB would raise rates, but only as far as Rome, Madrid, and Athens could absorb. Lagarde called this “fragmentation protection.” Markets had a more precise term: fiscal dominance. The ECB retained its formal independence. Its functional independence was already gone.

· · ·

The arithmetic is not contested. A state’s debt ratio stabilizes when its nominal growth rate exceeds the interest rate on its debt, provided its primary budget is roughly balanced. When the rate falls below growth, the ratio shrinks. When it exceeds growth, it compounds upward.

The mechanism is visible in Italy’s own recent numbers. Real GDP growth is projected at 0.4 percent in 2025, with the primary surplus rising to 0.9 percent of GDP. The debt ratio is set to reach 137.2 percent of GDP by 2027, as the expected primary surpluses remain insufficient to offset the impact of debt-increasing interest-growth-rate differentials. In plain terms: Italy is running a surplus, growing modestly, and its debt ratio is still rising. The gap between what it earns and what it owes in interest is the problem — and that gap is determined in Frankfurt, not Rome. Economy and Finance

Italy cannot afford sustained high interest rates. Not because of fiscal recklessness — the primary surplus demonstrates the opposite. The problem is structural: at this level of debt, even moderate rates become unsustainable. The country’s fiscal trajectory depends not on its own policy choices, but on the ECB’s.

Frankfurt knows these numbers. Italy, Spain, France, and Greece together account for roughly 60 percent of Eurozone GDP. A rate policy that destabilizes this group destabilizes the currency. The ECB operates under an implicit ceiling: rates may not rise beyond what the most indebted large economies can service. No government issues instructions to the central bank. Fiscal reality does.

· · ·

This was not supposed to happen. The Eurozone was designed as a monetary union without a fiscal union, on the theory that market discipline would substitute for fiscal transfers. The theory was coherent. The outcome was fiscal dominance.

The United States issues Treasuries — a single globally liquid asset, underpinned by structural dollar demand that exists regardless of yield. Every Eurozone state issues its own bonds. German Bunds are safe assets; Italian BTPs are not. A rate increase hits Italy harder than Germany. The ECB must manage that spread as a side condition of every monetary policy decision it makes.

Mario Draghi’s “whatever it takes” in July 2012 was not a free choice. It was the only available response to a union whose architecture had produced an institution forced to rescue the currency by redefining its own role. The Outright Monetary Transactions programme was never activated. The announcement was enough. Italian yields fell from near seven percent to under three percent within months. A decade of near-free government financing followed, during which debt ratios continued to climb. Lagarde’s successor instrument — the TPI, with no defined activation threshold and a trigger condition Frankfurt alone determines — replaced it as the standing backstop. The program has never been used. Its existence is the policy.

· · ·

Fiscal dominance is stable under three conditions: low inflation, sufficient growth, and sustained market confidence. None is guaranteed.

If inflation stays above two percent, the ECB must choose between raising rates and triggering debt crises across its largest economies, or tolerating inflation and destroying its credibility. Either damages the institution.

If growth stays low, debt ratios rise even at low interest rates. Italy’s economy grew just 0.7 percent in 2024, and growth is projected to moderate further to 0.5 percent in 2025. At that pace, even a modest rise in borrowing costs outpaces the economy’s capacity to grow its way out. International Monetary Fund

If markets lose confidence, yields rise faster than the ECB can respond. The OMT and TPI work because markets believe they will. If that belief is tested simultaneously across Italy, Spain, and France — several trillion euros in outstanding debt — the credibility of unlimited intervention collapses. The ECB can plausibly rescue Italy. Three major economies at once is a different claim.

THE VERDICT

The ECB does not freely set interest rates. It sets the highest rate that Rome, Madrid, and Paris can service without triggering a crisis, and calls the result monetary policy. Italy overtaking Greece is not a headline. It is a data point in a longer series: debt ratios that compound quietly, privatization targets that are missed, primary surpluses that are real but insufficient. Fiscal dominance is not a governance failure — it is the structural consequence of building a monetary union without a fiscal one, then allowing debt ratios to accumulate for two decades while the institution nominally responsible for price stability became the union’s financial backstop by default. The arrangement holds until it is tested. The question is not whether that test is coming. It is what breaks first when it arrives.

Three weeks later, Eurostat published its annual trade figures. The EU’s goods deficit with China had reached €360 billion in 2025 — up 18 percent from the year before — as Chinese manufacturers, shut out of the American market by Trump’s tariffs, redirected their exports westward into a bloc that still applies duties of two to three percent on most of what Beijing sells. Europe condemned the flood. It kept buying.

These two facts do not sit in tension. They are the same mechanism, visible from different angles.

· · ·

The EU’s official position is that it can manage three things simultaneously: security partnership with Washington, economic engagement with Beijing, and enough strategic independence to act in its own interest when those partnerships conflict. Brussels calls this “strategic autonomy.” The concept has been official doctrine since roughly 2016, repeated in every major policy document since.

The problem is not that the goal is wrong. The problem is what it costs — and who would pay it.

Genuine autonomy in defense requires Europe to build capabilities it currently buys from America: advanced fighter integration, missile defense architecture, satellite intelligence. The US supplied 64 percent of European NATO members’ weapons imports between 2020 and 2024. The EU fields over 170 different weapons systems; the United States fields 30. Closing that gap requires decades of consolidated procurement, which requires surrendering national industrial preferences, which requires telling France that its defense sector competes with Germany’s, and telling Poland that buying American F-35s — which it desperately wants for geographic reasons — undermines a collective goal it also endorses. The cost is not financial. It is political. It lands on specific voters in specific countries before any security benefit materializes.

Genuine autonomy in trade requires Europe to accept that replacing Chinese inputs — in batteries, electronics, solar components, pharmaceuticals — means paying more for them, at least during any transition. Chinese goods currently restrain European consumer price inflation in measurable ways. Removing that pressure without a domestic supply alternative is a tax on households. Governments that impose it lose elections to parties that promise to remove it. This is not speculation; it is the revealed preference of every European electorate that has been asked to absorb the cost of economic restructuring in the past decade.

The trilemma, then, is not geometric. It does not arise because the three goals are logically incompatible in some abstract sense. Japan and South Korea maintain deep security ties with Washington while selectively decoupling from Chinese supply chains in specific sectors. Partial autonomy is not impossible. What makes it persistently undelivered in Europe is a distributional problem: the costs of building it are concentrated, immediate, and politically attributable, while the benefits are diffuse, delayed, and impossible to assign to any government that will face an election before they arrive.

· · ·

This explains a pattern that otherwise looks like incoherence.

The EU’s new Readiness 2030 plan excludes American contractors from its €150 billion SAFE defense facility and requires a minimum 65 percent EU content threshold for SAFE-funded procurement. These are the right policy directions. They are also in direct conflict with the immediate procurement reality: European production lines cannot currently supply what European militaries need on the timescale that a deteriorating security environment demands. Poland, which shares a border with a country at war, spent more on US weapons between 2022 and 2024 than any other European nation — because Polish security planners need equipment that exists, not equipment that will be manufactured in a consolidated European defense industrial base that does not yet exist. Warsaw endorses strategic autonomy. It also buys F-35s. Both positions are rational given the time horizon each is optimizing for.

The China dynamic is structurally parallel. The Commission imposed tariffs on Chinese electric vehicles in 2024, presenting them as a defense of European industrial competitiveness. Within months, Brussels was negotiating the minimum pricing mechanism it had previously rejected — because the alternative was blocking the EU’s own climate transition, for which European manufacturers do not yet produce affordable vehicles at sufficient scale. The tariff was designed to protect European industry. The negotiation protected European consumers and climate targets instead. Neither side was wrong. The cost of choosing one was borne by the other, and no political majority existed to impose it.

What makes this structurally durable rather than a temporary misalignment is the time gap between cost and benefit. A government that genuinely reduces defense dependence on Washington will spend more, accept inferior capabilities in the short term, and antagonize an ally — for gains that materialize in fifteen years, after three election cycles. A government that genuinely reduces economic dependence on China will raise consumer prices and disrupt supply chains — for resilience gains that are invisible until the crisis they were designed to prevent. In both cases, the voter who bears the cost and the voter who receives the benefit are not the same person, and often not the same generation.

· · ·

Strategic autonomy survives as doctrine precisely because it does not require this choice to be made now. It is, in that sense, a perfect political instrument: ambitious enough to signal seriousness, indefinite enough to defer the bill. Every planning cycle produces a roadmap. Every roadmap extends the timeline. The dependencies deepen in the interval.

The irony that Brussels will not articulate is this: the United States spent years demanding that Europe pay for its own defense. Europe is finally doing it — and directing the money partly at excluding American contractors, while remaining operationally dependent on American systems for which no European replacement is scheduled. Trump demanded NATO pay its dues and received, instead, a rearmament program designed to circumvent American leverage. The relationship is loudly being renegotiated. The dependency is quietly not.

France understands strategic autonomy as industrial policy: the defense sector as a European champion, procurement as the instrument. Germany has historically understood it as a reason not to choose between Washington and Beijing. Poland understands it as rhetoric that should not interfere with the F-35 contract. All three positions are internally consistent. Together they produce a consensus document and divergent national budgets — which is exactly what thirty years of European strategic autonomy looks like in practice.

THE VERDICT

Strategic autonomy is not structurally impossible. It is politically undeliverable — and the distinction matters. The reason Europe does not build genuine independence from Washington’s security architecture or Beijing’s manufacturing base is not that the goals conflict in some abstract geometric sense. It is that the transition imposes concentrated costs on identifiable voters before it produces diffuse benefits for future ones. That is a problem democratic systems solve badly in general, and solve worst when the costs are international and the electorate is national. The €360 billion deficit with a designated systemic rival and the 170 weapons systems that 27 governments cannot consolidate are not failures of ambition. They are democracy producing exactly the outcome its incentive structure predicts.