Europe Just Had Its Best VC Quarter in Four Years. It Still Loses.

Europe-based startups raised roughly $24 billion in the second quarter of 2026. That is up about a third from the previous quarter and roughly two-thirds higher than the $14.4 billion raised in Q2 2025, and it is the strongest venture quarter the region has posted in four years.^[1] The UK alone took $10.4 billion of it, its third-largest quarter on record and within about $400 million of the 2021 peak.^[1] Germany came in at $3.2 billion, France at $2.4 billion, Sweden at $2 billion.^[4]

Every European tech newsletter I read framed this as a turning point. I don't think it is. In the same three months, startups in the US and Canada raised $137.2 billion.^[3] Global venture funding hit $205 billion, the second-largest quarter ever recorded.^[2] Europe's share of that is under 12%. A year ago Europe was raising $14.4 billion in a smaller global quarter. In percentage terms the region barely moved. In absolute dollars, the gap between Europe and North America got bigger, not smaller.

That is the thing about a rising tide. It lifts the boat you're in and the boat you're racing against, and if the other boat is five times the size, the tide is not your friend.

The money is real. The breadth is not.

Look under the headline and the quarter gets thinner fast. Four companies raised rounds of a billion dollars or more, and those four deals accounted for 25% of all startup investment in the region in Q2.^[1] They were Isomorphic Labs, the Alphabet-owned drug discovery company spun out of DeepMind; Stegra, the Swedish green-steel manufacturer; Neura Robotics, the German humanoid and industrial robot company; and Ineffable Intelligence, an AI lab founded by former DeepMind researchers.^[1] Three of the four are AI or AI-adjacent. One of them is majority-owned by Google.

Funding to Europe's AI companies passed $10 billion in the quarter, the largest quarterly figure the region has ever posted.^[1]That's the whole story, basically. Take AI out and Europe had a normal quarter. Late-stage funding hit $12.1 billion, up 90% year over year, which sounds like the growth-capital problem is finally getting solved.^[1] But that late-stage number is the same four megarounds wearing a different hat.

Meanwhile the deal count fell, and it fell mostly at seed.^[1] That is the tell. Capital concentrating into a handful of enormous rounds at the top while the number of new companies getting their first real check goes down is not a healthy ecosystem broadening out. It is a K-shaped market. It is what happens when a small number of investable AI assets absorb everything and the rest of the market gets picked over.

The chart Europe doesn't want to look at

Here is the comparison nobody in the celebratory coverage ran. Same quarter, same data source, Europe against North America.

Crunchbase noted that a single giant financing for Anthropic accounted for roughly half the North American quarterly tally.^[3] Fine. Strip it out and North America still raised something in the neighborhood of three times what all of Europe did, from a talent pool that is not three times better. This is not an engineering problem. It never was.

Defense is the one genuinely new thing

I want to be fair to the bull case, because there is one, and it isn't AI. It's defense.

European defense, security and resilience startups raised $8.7 billion across 2025, up 55% from 2024. Defense now accounts for roughly 10% of all European venture funding, versus less than 1% before 2020.^[6] Munich-based Helsing closed a $1.8 billion Series E at an $18 billion valuation, the largest defense-tech startup round in European history, with investor demand reportedly exceeding the available allocation.^[5] Quantum Systems, another German defense firm, raised $1.2 billion at an $8 billion valuation.^[5]

This one is structurally different from the AI rounds, and here's why I care about the distinction. Defense procurement in Europe is sovereign by law. A German drone company selling to the Bundeswehr cannot simply relocate to Palo Alto and keep the contract. The customer is the state, the state is European, and the moat is regulatory. Helsing reportedly remains predominantly European-owned after the round.^[6] That is a category where Europe can build $50 billion companies that structurally cannot leave.

Compare that to a European AI infrastructure company. Its customers are global, its investors are increasingly American, its compute is rented from a US hyperscaler, and its best exit is a Nasdaq listing. Nothing pins it down. Defense tech is the only major European venture category with real gravity, and it exists because Vladimir Putin created a procurement budget, not because Brussels designed a policy.

The exit problem is the actual disease

Funding is the symptom people count because it's easy to count. The disease is exits.

Across EMEA, M&A has accounted for more than 85% of venture-backed exits over the last five years, and the number of EMEA venture-backed companies going public sits at a decade low.^[7] Read that again. In the region with the second-best engineering talent on earth, the default outcome for a successful startup is getting bought, usually by a strategic acquirer, often American.

And when European companies do go public, look where. Bending Spoons, the Milan software company, listed on Nasdaq on July 1, 2026, raising about $1.68 billion and closing nearly 40% above its IPO price.^[13]Klarna, Sweden's fintech, listed on the New York Stock Exchange in September 2025.^[13] Nasdaq just posted the strongest first half in US exchange history, raising $129.3 billion from new listings.^[13] Meanwhile nine European startups worth a combined $12.3 billion or more announced plans in 2026 to merge with US-listed SPACs (special purpose acquisition companies, shell companies that take a private business public by merger), the highest annual tally since 2021, concentrated in nuclear, quantum, space and defense.^[8]

Think about what a SPAC actually signals. These are capital-intensive European deep-tech companies choosing a structurally worse, more expensive, more dilutive route to a US listing over a clean listing at home. They are not doing that because they love SPACs. They are doing it because the European public markets cannot price them, cannot give them index inclusion, and cannot give them the specialist tech investors and trading liquidity that a Nasdaq listing does.^[13]

Europe has the money. It just refuses to invest it.

The bit that makes me genuinely angry is that this is not a wealth problem. Europe is rich. It just doesn't put its own money anywhere near its own companies.

Roughly 0.01% of European pension fund assets under management is invested in European venture capital, about a third of the equivalent US rate. Matching US allocation levels would add an estimated $210 billion to European tech over the next decade.^[9] Atomico puts the cumulative shortfall at $375 billion of underfunding over the past ten years, with a minimum of $1 trillion more needed over the next ten just to stop the gap from widening.^[15] Almost half of the funding for European late-stage startups now comes from US and Asian investors.^[9]

That last number is the whole thesis in one line. Europe seeds the company with European money. The company hits Series C. An American fund writes the growth check, takes the board seat, and quietly starts talking about a Nasdaq listing in four years. Europe pays for the risky part and America owns the compounding part. It's a terrible trade and it happens because a Dutch pension fund would rather hold German bunds than a Dutch Series C.

The fragmentation makes it worse. In the US, the top 10 venture funds capture around 40% of total capital commitments, which is what a winner-take-most market looks like when it works. In Europe the top 10 capture roughly 20%.^[9] Sounds more democratic. It just means no European fund is big enough to lead a $500 million growth round on its own, so the round goes to whoever can, and that firm is on Sand Hill Road. US companies are twice as likely to raise a $50 million-plus round in the first place.^[9]

What would actually change my mind

The interesting policy thing on the table is EU Inc., the so-called 28th regime. The European Commission published a legislative proposal on March 18, 2026 for a single, optional, harmonized EU-wide company form covering corporate, insolvency, labour and tax rules, with registration through a central interface, a 48-hour deadline and a €100 cost ceiling.^[11][12] Parliament and member states are working on it now, aiming to agree a final text by the end of 2026, with first registrations potentially available in Q1 2027.^[16]

If it lands, it kills a real tax on European startups: the fact that raising across 27 legal systems is genuinely harder than raising across 50 US states. That's worth something. But incorporation was never the reason Klarna listed in New York. Fixing the paperwork does not create a domestic buyer for a $30 billion European technology company.

Three things would make me revise this piece.

1. Pension allocation mandates with teeth.Not a voluntary industry compact. An actual regulatory change that moves European pension and insurance capital into European growth equity at scale. The UK's attempts at this have been directionally right and far too small. Move the 0.01% to 0.1% and the late-stage problem disappears within a cycle.
2. A European exchange that a $20 billion tech company would actually choose.Deep specialist tech coverage, real index inclusion, and enough liquidity that a founder isn't taking a valuation haircut for staying home. Until a Bending Spoons-scale company picks Amsterdam or Frankfurt over Nasdaq and doesn't regret it, nothing has changed.
3. Three consecutive quarters of broadening, not concentration. Rising seed deal counts. More companies raising $50 million-plus rounds, not fewer companies raising bigger ones. That would tell me the ecosystem is deepening rather than a few AI names absorbing a global bubble.

So what do you actually do with this

If you're an investor: the European exposure worth owning is the stuff that can't relocate. Defense primes and the startups selling into them. Industrial decarbonization with physical plant and EU policy support. Regulated fintech infrastructure. Buy the gravity, not the talent, because the talent is a globally traded commodity and you will not capture its upside from a European listing.

If you're a founder: this quarter changes nothing about your Series C. The growth check is still probably American, and the person writing it still probably wants you in New York in five years. That is not a moral failing, it's a rational response to where the capital and the exit liquidity live. Optimize accordingly, and stop expecting a headline number to fix it.

The $24 billion is real. The 66% growth is real. The AI money flooding into Europe is real, and Helsing at $18 billion is a genuinely great outcome. But a quarter where four deals are a quarter of the money, seed deal counts are falling, IPOs are at a decade low, and your best companies are reaching for US SPACs is not a turning point. It is the same structure, with more money in it. Europe didn't fix the leak. It just had a good month for water.