The World Without the UPC: How a Successful Court Could Still Disappear

The most uncomfortable thing one can say about the Unified Patent Court is not that it is becoming too German. It is that it may never have been built, and is still not quite treated, as a European court.

That is unfair, and the unfairness is the point. Legally, the UPC is a common court of the Contracting Member States; its Agreement says as much, its judges act on that basis, and its panels are multinational by design. The Court of First Instance comprises a central division — with its seat in Paris and sections in Munich and Milan — alongside local and regional divisions, and Article 8 UPCA guarantees that first-instance panels are composed across nationalities.

But the legal form of an institution is not the whole of its life. A court can be European in its founding texts and national in its reflexes; common by treaty and concentrated by practice; equipped with several addresses and possessed of only one real centre of gravity.

This is why the current argument about case distribution deserves better than the shrug it usually receives. It is not an administrative embarrassment, nor a provincial quarrel about who gets the work. It may be the first genuinely constitutional debate of the UPC era — and the Court would be wise to treat it as one.

An earlier round of that debate, about appeal rates, established a useful negative: Europe should not turn the UPC into a league table. Appeal statistics are not a scoreboard ranking Munich against Mannheim, Düsseldorf against Paris, The Hague against Milan, and a Court of Appeal is not a referee of national pride. All true. But declining to keep score does not dispose of the distribution question; it sharpens it. Precisely because Europe is not a competition between jurisdictions, one is entitled to ask whether the UPC is doing enough to spread its authority across the Continent it claims to serve. The worry is not that Germany has done too well. It is that Europe has not yet done enough.

When Success Becomes Concentration

Call the future worth avoiding the monocentric court. Germany has not captured the UPC, and the suggestion that it has would be both wrong and ungrateful. German judges, German divisions and German practitioners have done a great deal to make the Court credible: they brought speed, procedural discipline and a seriousness without which the UPC might have remained a handsome machine admired chiefly by the people who designed it. That should be said plainly, and not as a throat-clearing courtesy before the real argument. It is part of the argument.

The problem is not that Germany has done too well. It is that the rest of Europe has not yet done enough. The risk the acronym describes would attach to whichever national culture happened to become the system's default — and it has attached, for now, to the German divisions only because they succeeded first and most visibly. Redistribution, on this reading, is not a penalty imposed on excellence. It is the institutional price of continuing to call the UPC a European court.

The debate beneath the debate

On the surface, the dispute is about filings. Users choose Germany because they trust German patent litigation: they know the judges, the rhythm, the style, the remedies, and they can translate decades of pre-UPC experience into the new system with little friction. There is nothing irrational in that. A claimant selecting a forum is enforcing a patent, not writing a love letter to European integration, and we should not pretend otherwise.

The deeper question is whether user preference, left to itself, can build a European court.

One answer is that the market has already spoken. Users choose certain divisions because they trust them; concentration follows from informed choice. There is something to that. But European courts are not private marketplaces, and judicial legitimacy is not a commodity. Much of European integration has consisted, precisely, in building institutions designed to resist the spontaneous concentration of power and influence — not to ratify it.

The question is therefore not whether users should be free to choose. They should. The question is whether their choices, taken together, can supply a constitutional theory for a common court. That is a different inquiry entirely.

Markets reveal preferences. They do not determine how authority ought to be organised.

The distinction is hardly unique to patent litigation. European institutions are often created precisely where market outcomes and constitutional choices diverge. The location of the Court of Justice, the European Central Bank or the European Parliament was never left to market selection alone. Judicial authority belongs to the same category.

The distinction is not merely historical. The United States Constitution explicitly links intellectual property to the promotion of scientific and economic progress — Congress is empowered to secure exclusive rights in order to promote the progress of science and useful arts. European law adopted a different vocabulary. Article 17(2) of the Charter of Fundamental Rights simply states that intellectual property shall be protected. The emphasis is not on economic optimisation but on legal protection within a broader constitutional order.

The UPC was born from that tradition. It was not created to optimise litigation flows, any more than the Court of Justice was created to maximise judicial throughput. It was created to build a common legal space.

If it can, the logic runs in one direction only. The most trusted divisions attract the most cases; the most cases yield the most first-instance decisions; those decisions supply most of the material the Court of Appeal works with; and the doctrine that emerges quietly takes on the procedural instincts of the busiest venues. No one needs to behave badly for this to happen — and that is exactly what makes it serious. The claimant acts rationally, the defendant adapts rationally, the judges work conscientiously, the Court of Appeal harmonises, commentators tally, the market applauds, and a court that no one set out to narrow becomes narrower than its founding promise. A monocentric court is rarely created by design. It emerges through the accumulation of individually rational choices.

Numbers are not everything, but they are not nothing

The figures call for caution. The Court is young, the samples are fragile, and the totals shift depending on what one counts — infringement actions, preliminary injunctions, revocation actions, counterclaims, procedural orders, appeals, or decisions on the merits. The direction of travel, however, is hard to miss. A June 2026 analysis marking the Court’s third anniversary placed the German local divisions at 70% of preliminary-injunction filings and 76% of infringement main actions, with Munich the most frequently chosen division in both categories and The Hague and Milan trailing at roughly 10% of preliminary injunctions each. Other readings of the data point the same way while suggesting movement at the margins: one analysis put German-based divisions at 66% of year-to-date infringement complaints through April 2026 — down from 78.4% in the second half of 2025, and as low as 50% in the single month of April. An earlier estimate had the German divisions at 77% of infringement actions filed to that point.

These numbers prove no bias. They do not show that the German divisions are too strong, too claimant-friendly or too national. They establish something more modest and more consequential: the Court currently has a centre of gravity. That is neither surprising nor illegitimate. The question is whether that centre of gravity gradually becomes a shared European one. They decide where legal questions first arise, where procedural disputes are tested, where evidence is fought over, where confidentiality and claim-construction practice take shape, where injunction doctrine is refined — and, in the end, where the Court of Appeal sources the raw material from which it builds common law. A young court makes its law wherever its cases go. The faint downward drift in the German share is therefore not a footnote to be brushed aside; it is the first sign that the trajectory is not fixed, and that a more even distribution is achievable rather than utopian. Distribution is not a cosmetic concern. It is law-making before the doctrine has a name.

Europe has been here before

The UPC is not Europe’s first attempt at a common patent jurisdiction. It is the latest survivor of a long and instructive failure rate, and the lineage is worth recalling accurately. The 1975 Community Patent Convention, reworked in 1989, never entered into force. The European Patent Litigation Agreement — launched in 1999 under the name European Patent Litigation Protocol and steered by a Working Party on Litigation under the auspices of the European Patent Organisation — produced draft texts in 2002 and 2003 before being abandoned in 2007 amid doubts about its compatibility with Community law. And the European and Community Patents Court, the most integrated design of all, was declared incompatible with the EU Treaties by the Court of Justice in Opinion 1/09. The UPC Agreement, signed in 2013 and rebuilt around participating EU Member States, is the answer Europe finally found; the EPO’s own published travaux préparatoires trace the connection back through these earlier projects.

The history matters because each attempt leaned toward a different model. One is the international-technical model — efficient, expert, pragmatic, at home in the world of the European Patent Office and, like the EPLA, open in principle to non-EU states. Another is the EU-constitutional model — integrated, judicially disciplined, alert to the autonomy and primacy of Union law, and the model Opinion 1/09 effectively made mandatory. The third is the national-litigation model — familiar, trusted, procedurally mature, and not necessarily European in spirit. The UPC is a negotiated compromise among the three. The compromise is what made the Court possible; it is also what makes it fragile. The danger today is not that a German model has defeated a European one. It is that the UPC may never have grown a European model strong enough to stop the most successful national model from becoming the practical default.

Does a European Court Need a Centre?

The UPC already has a centre. The central division sits in Paris, with sections in Munich and Milan. That was a deliberate constitutional choice. It is worth asking whether it is doing constitutional work.

European institutions have generally found their footing once they acquired a visible centre. The EPO settled in Munich. The Court of Justice in Luxembourg. The ECB in Frankfurt. None became less European for it. A centre, in each case, helped turn an international arrangement into a stable institutional reality.

The UPC was designed differently — as a genuinely distributed court. That design reflects something real about Europe: authority should be shared. But shared authority and absent authority are not the same thing. The question is not whether the UPC should become more German or more French.

The question is whether it has developed a European centre of gravity at all.

At present, the answer is not obvious. Much of the Court's practical energy comes from Germany. Much of its constitutional legitimacy derives from Luxembourg. Its formal centre sits in Paris. Milan is growing as a judicial actor. The Hague continues to attract cases. None of this is a problem in itself. The difficulty is that these elements do not yet add up to a coherent institutional story.

Users know where to file. It is less clear that they know where Europe is.

A court can function without a dominant city. It is less obvious that it can endure without a recognisable constitutional centre.

That may sound abstract. It is not. Institutions hold when their users understand them not only as procedural mechanisms but as genuine legal communities. The risk for the UPC is not that German divisions keep succeeding. The risk is that their success becomes the Court's only legible narrative — and that filing patterns come to substitute for institutional identity.

The lesson of the EPO is not that Europe needs another Munich. It is that Europe needs a convincing answer to where its common institutions actually reside.

The UPC may find that answer in Paris, in Luxembourg, in a more balanced network of divisions, or in some combination of all three. But if it does not find one, others will eventually start looking for an alternative design.That is where the world without the UPC begins. Perhaps the ultimate test of a European court is not whether everyone knows where to file, but whether everyone would know where to rebuild it.

What redistribution should mean

It is worth being precise about what the argument is, because its opposite is easy to caricature. The case for redistribution is not that Germany has too many cases; it is that a European court needs a European ecology of law-making. It is not that users should be steered away from the German divisions; it is that the Court must give them reasons to trust more of itself. It is not that Paris, Milan or The Hague deserve protection; it is that no division of a common court should be allowed to become institutionally decorative. The point is not to make the German divisions matter less, but to make the rest of the Court matter more. The objective is not equal distribution. It is sufficient distribution. A European court does not require symmetry. It requires participation.

That distinction is not a rhetorical nicety — it is the whole game. If the debate curdles into Germany-versus-the-rest, it will fail both intellectually and politically, and it will also be unjust. Germany cannot be asked to carry the UPC and then be blamed for carrying it too visibly. The right question is not whether German success should be restrained, but whether European underuse elsewhere should be read as a constitutional warning rather than a marketing problem. Not every institutional question can be reduced to attractiveness metrics. Courts are not platforms, and legitimacy is not market share.

It should not mean crude quotas, nor a registrar shuffling cases around the Continent like misplaced furniture, nor any weakening of the divisions users currently trust. Punishing the parts of the Court that made the system attractive would be a strange way to make it more European. The sensible measures are more precise, and most of them are available now.

The first is information. The Court should publish, at regular intervals, a serious European Distribution Report — not a brochure and not a ranking, but an institutional document that breaks the docket down by action type, remedy sought, technical field, language of proceedings, value in dispute, domicile of the parties, division of origin, panel composition and appellate treatment. The aim is not to crown a winner but to see, honestly, where the Court’s law is actually being made.

The second is a habit of mind: a distribution-impact lens on institutional choices. When practices are settled, panels organised, languages designated, guidance issued, judges allocated or case law communicated, one question should be asked aloud — does this make the Court more European in use, or merely more efficient where it is already strongest?

The third concerns the bench itself. Article 8 already gives first-instance panels a multinational composition, but a formal guarantee is not the same as a visible one. In suitable cases, particularly in high-volume divisions, the non-national judge could more often serve as judge-rapporteur or take a more prominent role in the conduct of proceedings — not for the sake of symbolic rotation, but to make plain that even where a case is filed in a national location, the authorship of the decision remains European.

The fourth is predictability. Users do not choose Germany because it is Germany; they choose it because they believe they know what will happen there. The answer is not to moralise that choice but to make the alternatives equally legible — through Court-wide transparency on timetables, hearing structure, preliminary indications where appropriate, confidentiality practice, evidence management, amendment practice and security for costs. Predictability should be a property of the Court, not a feature of some of its rooms.

The fifth is language. English already dominates UPC practice — by one account it accounted for more than two-thirds of Court of First Instance filings through April 2026— and making it practically available across more divisions, without displacing national languages, would remove one of the real obstacles to using non-German venues. The goal is not linguistic uniformity but equal accessibility.

The sixth is a soft anti-concentration principle for coordinated litigation campaigns. Where the same claimant or corporate group brings several related actions against the same defendants in a short window, the system might at least ask whether they all need to land before the same national cluster of divisions. The current Agreement does not permit forced reallocation: Article 33 gives parties defined routes for bringing actions and allows referral to the central division only in specified situations, including by agreement of the parties. But guidance, case-management incentives and party-agreed referrals can already shift practice without pretending the Court holds powers it does not.

The seventh is to take the Article 87 review seriously, and early. Article 87 provides for a broad consultation with users on the Court’s functioning, efficiency, cost-effectiveness and the confidence it inspires, to be held seven years after entry into force or once 2,000 infringement cases have been decided, whichever falls later. That review should have a distribution chapter — and the Court should not wait for the formal milestone to start gathering the data. A shadow review can, and should, begin now.

What practice can do, and what only a treaty can

Keeping these categories distinct is itself a matter of legal hygiene, because a redistribution argument that overreaches will be dismissed as quickly as it is made. Better data, more candid communication and a more visible use of multinational panels require no amendment of anything. Broader availability of English is largely a matter of national and institutional choice, depending on language designation. Encouraging party-agreed referrals to the central division where Article 33 already allows them requires no fiction about the Court’s powers.

Other ambitions are different in kind. A more flexible jurisdictional role for the central division, a revised regime for declarations of non-infringement, or a genuine mechanism for handling coordinated litigation campaigns would, on any honest reading, require amendment of the UPCA itself. They should not be smuggled into the Rules of Procedure if the Agreement does not support them. A court does not become more legitimate by quietly stretching its own legal basis — which is exactly why the redistribution debate should be institutionalised and argued in the open, rather than improvised through practice and presented after the fact.

The world without the UPC

A world without the UPC is not, in the likely case, a world in which the Court vanishes — that would be too cinematic, even for UGC. The plausible failure is quieter. The Court remains, the website works, decisions issue, conferences multiply, users keep filing, the Court of Appeal keeps harmonising, and by every visible measure the institution succeeds. What narrows is the mental map. The risk is that the practical image of the Court becomes narrower than its institutional design. The other divisions stay competent, respected and formally equal, but underused — and their want of cases becomes the reason not to send them cases, while their want of a track record becomes the reason they can never build one. The Court remains European in law and stops being European in the professional imagination.

That is the real concentration scenario: not national dominance, but a European failure to distribute trust.

The first is the managed UPC, and it is the one to aim for. The Court treats distribution as a legitimacy indicator rather than a public-relations irritant: it collects better data, publishes honest analysis, levels procedural predictability across divisions, makes multinational panels more visible, encourages party-agreed redistribution where the Agreement permits, and prepares the Article 87 review with seriousness. The German divisions remain strong — but their strength no longer defines the Court's identity.

The second is the concentration future. No one abolishes anything; no one declares failure. The German divisions go on dominating first-instance infringement work, the Court of Appeal stays European in composition while the raw material beneath it stays geographically concentrated, and the other divisions come to look like formal options rather than natural venues. The Court works — as a European remedy delivered through a national comfort zone.

The third is the post-UPC European Patent Court, the speculative scenario, though not an absurd one. If the UPC were ever judged structurally incapable of becoming sufficiently European in practice, Europe might attempt once more what it has attempted for half a century: a more openly European court, more tightly bound to the EU legal order, with a stronger central identity and a more deliberate geographical balance. The lineage even hints that such a design need not be confined to the EU’s current membership — the EPLA, after all, was conceived as open to non-EU EPC states. But Opinion 1/09 is a standing reminder that a European patent court cannot be drawn up as though EU constitutional law were a technical footnote. Difficulty, it should be said, has never stopped Europe from trying to build patent institutions. It has only delayed them, complicated them, and occasionally made them stranger than fiction.

The question is not whether Europe needs another court. The question is whether the Court it has already built can fully realise its European vocation.

Conclusion: not a league table, but not a map to ignore

The UPC was created to overcome fragmentation, not to reproduce it in a more sophisticated key; to build a common jurisdiction, not to deliver a European remedy mainly through a national comfort zone; to move past the old national reflexes, not to identify the most successful one and rename it Europe. Seen that way, the provocation earns its keep. PPerhaps the next phase of the UPC's success lies not in becoming less German, less French, less Dutch or less Italian, but in becoming more visibly European.

That is why redistribution matters — not because Germany must be restrained, not because France, Italy or the Netherlands must be compensated, and certainly not because the Court should be arranged like a diplomatic seating plan. It matters because a European court that cannot distribute its own authority will sooner or later discover that authority follows habit, and that habit is a poor substitute for constitutional design.

There is still time, and the very existence of this debate may be a sign of health: young courts need pressure, correction and imagination. But if Europe means to avoid this outcome, it should stop filing distribution under statistics and start filing it under legitimacy. Distribution is the map of the Court's future — and maps, unlike patents, cannot be validated country by country. The future of the UPC will ultimately depend on whether Europe continues to see it as a common court — and not merely as a successful one.