Thursday, November 20, 2025
Executive Summary
A Berkeley Center for Law & Technology panel of European UPC practitioners — Rajvinder Jagdev (Powell Gilbert), Marc Lauzeral (Schertenleib), Yasmine Azzaoui and Carl-Alexander Dinges (Bonabry), and US co-author Steven Carlson (Robins Kaplan) — examined the court’s first thirty months of caseload data, finding that German local divisions handle approximately 80 percent of infringement actions, that just over half of all UPC infringement claims result in a finding of valid and infringed patent, and that preliminary injunctions are resolved within an average of 96 days — a combination of speed, patentee-friendly outcomes, and in-proceeding patent amendment rights that has no equivalent in any other major jurisdiction.
Instructor(s)
Marc Lauzeral, Schertenleib Avocats
Carl-Alexander Dinges,Bonabry
Yasmine Azzaoui, Bonabry
Steve Carlson, Robins Kaplan
Rajvinder Jagdev, Powell Gilbert
Keywords
UPC caseload statistics 2024–2025 — infringement actions by local division • UPC preliminary injunction grant rate and timeline •UPC patent amendment in proceedings — auxiliary requests before the court • German local divisions forum shopping — Munich, Düsseldorf, Hamburg, Mannheim • UPC Rule 190 — specific disclosure of documents in patent proceedings • BSH Hausgeräte GmbH v. Electrolux — UPC long-arm jurisdiction over non-member states • UPC Court of Appeal overturn rate — first instance decisions reversed • UPC transitional period 2030 — exclusive jurisdiction over non-opted-out European patents • Ireland UPC accession — English-language common law division • Docket Navigator UPC dashboard — patent case statistics Europe • “what percentage of UPC patent cases are won by the claimant” • “how long does a preliminary injunction take at the UPC”
Legal Analysis
German Division Dominance, Caseload Measurement Deficiencies, and the Emerging Shift in UPC Forum Distribution
Two and a half years after the Unified Patent Court commenced operations on June 1, 2023, the court’s caseload remains heavily concentrated in Germany’s four local divisions — Munich, Düsseldorf, Hamburg, and Mannheim — with Munich alone recording 1,197 filed cases in Docket Navigator’s October 2025 dataset, a number that dwarfs Paris, The Hague, and every other non-German division. Dinges attributed this concentration to a structural feature of the UPC Agreement itself: when a member state has had more than fifty patent cases litigated nationally in the preceding three years, the Agreement entitles that state to two standing national judges on each local division panel, and “therefore in all those German divisions, you have panels with two national standing judges, which makes the predictability of who is going to be your judge very high, and there’s a big, just huge room for forum shopping.” Germany’s pre-UPC position as the dominant European patent litigation forum — itself a product of the national bifurcated system that adjudicates infringement and validity separately, a structural advantage for patentees — translated directly into this compositional benefit when the UPC launched.
The raw caseload figures are, however, systematically overstated by the UPC’s own case-counting methodology, which Lauzeral identified as a persistent distortion. Every incident proceeding, confidentiality request, and ancillary application filed in connection with a single underlying dispute generates a separate official count, so that a claimant asserting five patents in one action who simultaneously requests preliminary relief and confidentiality protection may generate fifteen or more counted “proceedings” from a single real dispute. The panel’s statistics, drawn from Docket Navigator data supplemented by the practitioners’ own analysis of published judgments and orders, attempted to collapse these inflated figures into unique dispute counts, though Jagdev was candid about the resulting uncertainty: the data “isn’t made available to practitioners or to members of the public” through the UPC’s own case management system, and the official annual report — the first edition of which covered the year ending December 2024 — itself relies on the same overcounting convention. The UPC’s migration in 2025 to a new case management system built by the European Patent Office is expected to improve data quality, but as of the panel date the market remained dependent on third-party analytics.
Lauzeral predicted that the German divisions’ numerical lead will narrow over time as practitioners in France and the Netherlands grow more comfortable directing matters to the UPC rather than parallel national proceedings. The dynamic is already detectable: Lauzeral reported a close to 20 percent decline in national cases filed before French courts in the preceding six to seven months, and cited a colleague’s estimate of a 60 to 80 percent decline in national filings in the Netherlands over a comparable period. Azzaoui added that the forthcoming end of the UPC’s transitional period — currently set for 2030, with a possible seven-year extension — will render the court’s exclusive jurisdiction over non-opted-out European patents mandatory, creating institutional pressure for more balanced distribution among divisions: “there is a need to have a more balanced system … it will be interesting to bring more cases before those local divisions that have not many cases right now.” The life sciences sector already illustrates this pluralism, with Lauzeral noting that approximately sixty pharmaceutical patent disputes have been distributed across ten different local divisions, including Lisbon, Paris, Düsseldorf, Milan, and the Central Division.
UPC Preliminary Injunction Grant Rates, Bond Requirements, and the 96-Day Resolution Timeline
Preliminary injunctions constitute approximately 20 percent of all UPC cases filed, and the panel’s data showed that just over a third of all PI applications are granted in full, with a further approximately one-third granted in part — producing a combined success rate of roughly two-thirds for applicants seeking at least partial preliminary relief. The average time from filing to a closed PI decision is approximately 96 days, a figure that Jagdev contextualised against the US experience: “in the US, if you hear about preliminary injunctions, it’s like Bigfoot — I’m not sure that I’ve seen anything other than pictures of them in a decade.” The procedural mechanics explain the speed: the PI written procedure collapses the standard eight-month written phase for merits proceedings down to two to three months, with defendants given four to six weeks to file non-infringement and invalidity arguments, meaning respondents must simultaneously defend PI and merits proceedings on compressed, parallel timelines. Jagdev summarised the systemic effect: “the system is very much weighed in favor of the patentee claimant at the moment, given the front-loaded nature of it.”
The Court of Appeal has reversed approximately 50 percent of first-instance PI decisions, a figure Azzaoui attributed to the court’s early-stage development of binding appellate precedent on the applicable legal standards. The primary grounds for first-instance refusal are insufficient certainty as to patent validity and absence of imminent infringement, and the local divisions have adopted divergent procedural rules for PI applications: the Munich Local Division limits applicants to three substantive arguments, the Paris Local Division requires a pre-hearing summary of submissions, and the Düsseldorf Local Division has held that one month of awareness of infringement is sufficient to establish the diligence required to obtain emergency relief, while the Mannheim Local Division has expressly departed from that position and accepted a two-month window. Azzaoui predicted that appellate harmonisation of these procedural divergences will reduce the Court of Appeal’s overturn rate as the court’s own case law matures.
On the question of security bonds — a structural barrier to preliminary injunctions in the United States, where bonds may reach tens or hundreds of millions of dollars — the UPC operates with considerably more judicial discretion. Dinges explained that the court will generally decline to require a bond where the applicant is plainly solvent and able to compensate any damages caused by an erroneously granted injunction, as demonstrated by originator pharmaceutical companies seeking to restrain generic market entry. Conversely, in NPE enforcement actions against large market participants, Dinges confirmed that bonds would be “evidently” required, and that the evidentiary predicate for setting bond quantum — production of evidence demonstrating the likely damages from the injunction — creates a tactical dilemma, because claimants must partially disclose commercial harm to resist a high bond, yet risk those figures being used against them. Dinges also confirmed that irreparability of the harm to the respondent is a factor in the balancing of interests on a PI application, and that an applicant who cannot demonstrate protectable commercial harm in the relevant territories may face a significant obstacle to obtaining interim relief.
Infringement Success Rates, In-Proceeding Patent Amendment, BSH Long-Arm Jurisdiction, and the Language of Proceedings
The UPC’s merits-stage outcomes data is among the most commercially significant statistics the panel presented. Just over 50 percent of all UPC infringement claims result in a finding that the patent is both valid and infringed, a rate that Carlson described as strikingly favorable by comparison with the US Patent Trial and Appeal Board’s attrition record: “it was about only about a third of patent claims were surviving the PTAB, and then you got to survive the district court and find infringement, so it was a very small percentage of people actually winning in the US, compared to this slide, which is over 50% of claimants were winning.” Dinges noted that this figure represents a meaningful departure even from Germany’s pre-UPC national practice, where a bifurcated system produced a relatively high infringement-finding rate but simultaneously a high patent-invalidation rate, with the combination often resulting in a net injunction gap. The roughly uniform 50 percent valid-and-infringed rate across Munich, Mannheim, and Paris suggests that the pattern is not an artifact of any single division’s judicial philosophy.
One structural advantage for UPC patentees that has no equivalent before most national patent courts is the ability to amend the patent in the course of proceedings, subject to the constraint that amendments must narrow the claims and may only be filed once — at the reply to the statement of defense. Dinges characterised this as a “one bite in the apple” opportunity; Lauzeral observed that the amendment right is particularly valuable in life sciences cases involving complex pharmaceutical applications with multiple claim hierarchies, dosage regimens, and indication-specific claims, because it allows a patentee to tailor surviving claim scope to the infringing act without losing the infringement finding. A question from a judge of the IP High Court of Korea prompted clarification that UPC amendments are narrowing in nature, subject to appellate review, and are adjudicated by the tribunal itself rather than remitted to the European Patent Office. The interaction between the in-proceeding amendment right and the jurisdiction extended by BSH Hausgeräte GmbH v. Electrolux AB, Case C-339/22 (CJEU 2025) — which allows a UPC proceeding to adjudicate infringement of a European bundle patent in non-UPC member states such as Spain — raises a novel unresolved question: if the validity of the patent as amended governs the Spanish infringement claim, but Spanish courts retain exclusive jurisdiction over the validity of patents validated in Spain, the UPC’s amended claim scope may not bind the Spanish court. Lauzeral acknowledged the uncertainty, offering a preliminary view that Spanish substantive law would govern the validity analysis for the Spanish territorial claim.
Language of proceedings has evolved from a predominantly German-language practice — the natural consequence of Germany’s early caseload dominance — to one in which just over half of all UPC proceedings are now conducted in English, a shift that the judges themselves actively promote to maximise court accessibility. Carlson observed that patent claim construction is an exercise of linguistic precision for which native-speaker fluency is not merely a convenience: “patents are really funny documents with very intricate languages … and when you construe claims, so much hinges on just these little nuances of language.” Dinges cautioned, however, that the UPC’s English is “European English,” and that judges sometimes ask parties to simplify their register; native English-speaking witnesses present a particular challenge because subtle linguistic nuance may not survive a panel of non-native speakers. Jagdev identified Ireland’s potential accession to the UPC as the single development most likely to transform this dynamic, noting that an Irish division would bring both a common-law tradition — including familiarity with Rule 190 specific-disclosure applications and the use of expert evidence — and a natively English-speaking bench, conditions that Carlson predicted would make it “a very hot jurisdiction” for US-side litigants.
Generated by AI based on the Interview/Transcript below.
Key Takeaways
- Claimants Win Over 50% of Merits Cases: Just over half of all UPC infringement actions result in a finding of a valid and infringed patent, a win rate that Carlson characterised as dramatically superior to US practice, where only a third of patent claims were surviving PTAB review before district court infringement analysis even began.
- PIs Resolve in 96 Days on Average: Preliminary injunction proceedings close in approximately 96 days from filing, with just over a third fully granted and a further third partially granted, creating meaningful leverage for claimants without the bond burden common in US practice — though Dinges confirmed courts may require substantial bonds in NPE-against-implementer scenarios.
- Munich Dominates but Faces Reversal Risk: Munich’s local division handles more cases than any other UPC forum, but Lauzeral cautioned that two out of three Munich first-instance decisions are overturned on appeal, compared to one out of three for non-German local divisions, reducing Munich’s net predictability advantage for claimants.
- BSH Long-Arm Jurisdiction Raises Amendment Conflict: The CJEU’s 2025 decision in BSH Hausgeräte GmbH v. Electrolux AB permitting UPC adjudication of infringement in non-member states creates an unresolved question when the UPC amends the asserted patent in proceedings, because national courts of non-member states retain exclusive jurisdiction over patent validity for their territory.
- Patent Amendment Is a Unique UPC Tool: The UPC allows a patentee to amend claims — once, at the reply-to-defense stage — within the infringement proceedings themselves, a right unavailable before most national courts that Lauzeral identified as particularly powerful in life sciences cases with complex claim hierarchies.
- German Caseload Lead Will Narrow by 2030: Azzaoui predicted that the end of the transitional period (currently 2030) will force a more balanced caseload across divisions, and Lauzeral reported that French and Dutch national patent filings have already begun declining materially as litigants migrate to the UPC.
- Official Statistics Remain Unreliable: The UPC’s case-counting convention inflates figures by treating each ancillary proceeding within a dispute as a separate case; Lauzeral noted that what the official data calls “1000 proceedings” may correspond to far fewer unique disputes, and practitioners currently rely on Docket Navigator and firm-level analysis in the absence of granular official data.
- English Now Majority Language; Ireland Is the Prize: Just over half of UPC proceedings are now conducted in English, and Carlson predicted that Ireland’s accession — if ratified — would make its common-law, natively English-speaking division “a very hot jurisdiction” for US and UK-side litigants.
- Court of Appeal Overturn Rate Declining: Azzaoui reported that approximately 50 percent of first-instance PI decisions are currently reversed on appeal, but predicted the overturn rate will fall as the Court of Appeal builds a coherent body of precedent on validity-certainty and imminence-of-infringement standards.
B-CLE Recording(CLE: FREE) | Youtube Recording | Resource(s) | Speaker Bio(s) & Contact Info
Download the interview/transcript here!
Interview/Transcript
This transcript was based on a conversation on December 4, 2025 at the 2025 Berkeley-GW UPC Institute. This panel on “Update and Statistics on the UPC: Who is using it and what is happening?” was given by Marc Lauzeral, Schertenleib Avocats, Carl-Alexander Dinges,Bonabry, Yasmine Azzaoui, Bonabry, Steve Carlson, Robins Kaplan, and moderated by Rajvinder Jagdev, Powell Gilbert.
Rajvinder Jagdev 00:23
Very good, so hi everyone. I am chairing the second panel of the day. My name is Raj Jagdev. I’m a partner of Powell Gilbert. We’re a boutique patent litigation firm in London. Thank you, London, Dublin, and Dusseldorf. I’m joined on this panel today. The panel is about UPC statistics and updates, and we have a distinguished panel of European practitioners, and of course, Steve at the end. So, to my right here, I have Marc Lauzeral from Schertenleib. Marc’s based in Paris and has a practice focused on cross-border patent litigation, SPCs, and trade secrets, he has experience with the French, German, and Italian divisions of the UPC, and of course, the French courts as well. Then next to Mark, we have Yasmine Azzaoui, and then Carl Dinges, both from Bonabry. Carl is based in Munich, and his practice is a focus on high-tech patent litigation and Sep Fran litigation, and Yasmine is based in Paris, and her practices focus on life sciences and pharmaceuticals. And then at the end we have Steve Carlson from Robin’s Kaplan. Steve’s based in Silicon Valley, and there’s a seasoned IP litigator assisting clients with IP disputes, both in the US and internationally, and Steve is the co-author of a book called Patents in Germany and Europe: Procurement, Enforcement, and Defense, together with Alexander Harguth, other co-author, who you’ll hear from later. So this panel is looking, is going to look at the stats that we have available about the UPC so far, but I’m going to start with a health warning, because we don’t have great statistics, so we have a comprehensive case management system, which collects a lot of data, but that data isn’t made available to practitioners or to members of the public. The UPC does publish an annual report, and it published its first report for the year ending 2024 in June of this year, and it is not available, is not available. So it’s something we’re working on. There is a limited set of data available, and there are other resources I’ll get on to. So, Docket Navigator, for example, has a great UPC dashboard, and Docket Navigator, you guys use quite widely already. It’s relatively new in Europe, but it is a great resource for certain UPC statistics, and a lot of firms like ours collate our own data, and so judgments and orders are public, and so by parsing those, you can develop your own data set, and so the slides and graphs I’ll be showing you today are in part derived from docket navigator data, in part derived from our own analysis, and therein comes a health warning, and so they’re as accurate as we can get them, but they shouldn’t be relied on there for informational purposes only. The UPC report is a fantastic resource. If you’re new to the UPC, it has a set of stats at the end, but it also has a summary of the key decisions up until the end of 2024 So it’s a good introduction to the course and can help you get up to speed, but of course the court has been very active since the end of 24 We’ve had a lot of decisions in 25 that have helped to shape the case law, and so you’ll want to top it up with with more current reports. One of the key issues, and Mark, I know this is a bugbear of yours, is the way in which the UPC counts cases, and that feeds into some of the stats that you’ll see presented. Do you want to talk a little bit about that?
Marc Lauzeral 04:09
You know, it’s actually really simple. Every time you, you, you have an order, or you are medicin in incident proceedings, it count as a proceeding. So, sometimes you’re going to get one case, sometimes you have, I know. Cases when they enforce five different patents. If you are filing for request for confidentiality information, so you could have five at the beginning. You had five five proceedings, which was counting for the invalidity for the infringement action, then five for the content left for infringement, then five for if you request for confidential confidence. confidentiality of information, so at the end of the day, you have this huge amount of number of, like, we have 1000 proceedings, but it does not mean that you have 1000 cases, so that’s that, where the struggle is.
Rajvinder Jagdev 04:54
That’s exactly right, and so we go to a lot of effort as practitioners to try and devolve down. To the number of unique disputes, and so the graphs that we’ll see today show the number of unique disputes as best we can, we can determine that. So I thought I’d start with a bit of a sneak preview of Docker Navigator, just to help you orient yourself to the extent that you want to find a useful data source comparable to what you have for us education, we’re not sponsored by Docker Navigator. They’re not funding this, but it is just a great resource. And so there is a UPC library that’s available, and you can see on the right of the screen the source of stats that are available. There’s a very good dashboard with graphs and statistics, and you can also get access to decisions and orders and opt-out information as well for the prosecutors in the room, so I thought the UPC report would be a good place to start, and so these two graphs are taken from that report, and they show the number of actions brought, and the graph on the left focuses on revocation and infringement actions, and then the average time to trial on the right, and what you can see from the trend is that the number of infringed cases has steadily increased month on month, and it shows just how popular the UPC is proving. It really has taken off in a remarkable way, and the court is quite busy, and we’ll go on to see some stats about which divisions are the busiest, but the time to trial slide on the right is quite interesting. So one of the key promises of the UPC is trial within 12 months, and what we can see from that from that slide is that the average time to trial for an infringement action is actually for 405 days, so it’s slightly over the 12 month limit, but within a an acceptable margin of error, and so by and large the UPC is delivering on that 12 month timeline. Whether that is still the case at the moment remains to be seen. The next version of this report we published later in 2026 and certainly our experience is that some of the divisions are busier than others, and perhaps aren’t quite getting to trial in 12 months, but Steve, I thought this would be a good opportunity to bring you in to talk about how this compares to litigation in the US, both in district courts and also the ICC, which I think is perhaps the most comparable timewise to the UPC.
Steven Carlson 07:27
Yeah, I think in the US the fastest you’ll get trial is the IPC. You’ll get your initial determination in about 12 months, and then your final determination about 16 months. So that’s the fastest you’ll get, which is roughly on pace with, with the UPC little slower, but if you go to district court for damages, etc. you know, really taxes, you’re into two years, and and it’s up from there. On my case now, it’ll be more or less a four year timeline. It had a PTAB challenge in the middle, which slowed things down, so really you’re looking at, you know, three to four years in the US compared with, you know, a 12 month time to trial in the UPC. It’s really it’s night and day, and like Wayne said, it’s on the development cycle, you know, it’s in, it’s in your product development cycle timeline. So I always tell clients to consider the UPC, because it’s so much faster.
Rajvinder Jagdev 08:22
So, the next slide is taken straight from Docket Navigator. So, this is quite current. It’s as of the end of October this year, and what it shows is the usage across divisions, and you can see that by quite a mile, the German divisions are the busiest, with Munich being by far the busiest. I think the number there is slightly blur, but it’s 1197 unique cases, or filed cases, not number of unique cases, but it correlates to also the largest number of unique cases filed in in Germany, and this is a great opportunity to bring in our European panel members, because Yasmine, Marc, you’re in, of course, based in France, in Paris, and Paris is also quite popular, but that the German divisions have really taken the lead. Do you have any perspectives as to why that is?
Marc Lauzeral 09:29
So, I think historically is due to the fact that French, and it’s going to be, I think, the same thing for the Netherlands, are really close in the judicial system to the UPC. The fact that you have more proceedings in Germany does not mean that the food is better there, or sometimes you may have local division, which may be a bit more proper, and in the south of Germany, but apart from that, what happened is when the UPC was launched, when our client came to see us, and they said, “Okay, what should I do? Should I launch before national proceedings or before the UPC? You will say, “Look, you. PC is going to be the same thing than French proceedings, is going to be one year and a half, so a bit longer, but one year and a half proceedings to get the first instance decision, two months for a PI. Well, you’re going to discuss validity and infringement. The only thing which will really change is the fact that you will not be able to file auxiliary request, so you’re going to say to the client, UPC, it will only give you uncertainty, because we don’t know where the things will go. It was the same in the with the Netherlands, but in Germany, I think the system is really different, as you get a bifurcated system when you’re going to deal with one proceeding only relating to validity, one proceeding only relating to infringement. The UPC offer a true alternative, and what we have seen more and more, the fact that in Germany you had a transfer from national cases to the UPC, you had seen the number of national cases decreasing, and now we didn’t have that in the first year, in France and the Netherlands, for instance, we, the UPC cases were additional layers new litigation and now we start to see that I think from the past six seven months when you have influence close to 20% decrease in the number of national cases filed before national court I heard from a colleague I don’t know if he’s that high but he told me like 60 to 80% decrease in the Netherlands and what I’m, and so now you start to have that, so I think the new proceedings of the Paris Local Division, and even the Ag Local Division, when discussing with Brickman, that Judge Brigman, that’s what he told me, they’re starting to get there, so I think you’re going to have a more equilibrate system with maybe still the Germany will be important, and I see it will also really depend on some, but we will talk about it later, I assume. Some industry, the Germany is better for friends historically, because you’re going to have more certainty on that case.
Rajvinder Jagdev 11:51
and that’s right. You’ve made a very interesting point about the transition from national actions to the EPC, and Carl, in Germany, also quite a few of the national court judges have moved over to the UPC, as that is that a fact that plays into the popularity of the German divisions.
Carl-Alexander Dinges 12:09
Yes, I think so. So, practically even before the UPC was established, but by far the most cases have been brought to Germany, and that was at that time, probably before, because, because of the impact, and the, it was just the biggest country, and we have this bifurcated system, of course, which is a benefit to the patentee, and when the UPC was established, there were four local divisions, which were just there in located in Germany, and all of those local divisions have been assigned with judges from Germany, and there is an quite interesting provision in the UPC agreement saying the constitution of the panel depends on the number of cases which have in the last three years been litigated in that specific country, and if that number is below 50, then there will be one national judge in the panel, and if that number is above 50, then there will be two national judges in the panel, and therefore in all those German division divisions, you have panels with two national standing judges, which makes the predictability of who is going to be your judge very high, and there’s a big, just huge room for forum shopping, because you, you know, who is going to be the judge deciding on the specific question of the case to quite an extent, and I’m currently, since there have been quite a lot of cases, this has changed because there are further chambers assigned to the to the local divisions, and but right at the beginning I think it was quite interesting that you could not on a case by case level say this is a case which I would like to bring to this or that local division, because I know that at least I know at least two judges of a panel of three legally qualified judges and plus one technically qualified judge.
Audience 14:28
Can the panel talk about the judges a little bit. Are there hundreds of them, 1000s of them, like in the US, where hundreds, where it’s very diverse, or is it more specialized in smaller groups?
Carl-Alexander Dinges 14:46
It’s quite specialized. So, there will be so right at the beginning, there were only at the local divisions, there was only one chamber with the national judge, or the two national judges. Being standing member of that of that chamber, and then there was a pool of the European European pool of from from all other nations, which would bring in the missing third or second and third legally qualified judge, and I think that pool is on the legally qualified members, is about 3044 now, 44 now. So it increased a bit, and in addition, there’s a pool of randomly assigned technically qualified judges, which is a bit larger, I think, but I don’t know the exact number, so it’s rather, yeah, you could know about some of those judges.
Rajvinder Jagdev 15:46
and of course, when you say randomly assigned technically qualified judges, they’re assigned based on their technical ability, but not necessarily based on nationality, which does play into the assignment of legally qualified judges to cases. Do you want to say?
Yasmine Azzaoui 16:02
just wanted to add something. Considering the concentration of the cases before the German local divisions, we need to consider also what will happen after the end of the transitional period. So, for now, there is a dual jurisdiction of the UPC and of national courts regarding European patents that have not been opted out, and there, so after the end of the transitional period, which should be in 2030 it can also be extended by another seven years. So we still don’t know what will happen, but after the end of the transitional period, the UPC will have exclusive jurisdiction on those European patents, which have not been opted out, and the new European European patents that have been granted. So, there is a need to have a more balanced system with, like, number of cases that is balanced between all those local divisions, and it’s.. it will be interesting to bring more cases before those local divisions that have not many cases right now. I’m not talking about France, because France is getting more and more cases, but more, for example, the Lisbon local division, the Nordic Baltic divisions, to like gain more experience and
Marc Lauzeral 17:21
maybe just a quick one on that as well. Also, extremely industry dependent, when you are looking at SAP, like I said, you have a lot in Germany. When you are looking at life sentences, you’ve got, I think, almost 60 patent cases, which has been litigated, and they are in front of 10 different local divisions. So, you have people in Lisbon, you have people in Paris, Dusseldorf, Milan Central, so it’s really industry specific
Rajvinder Jagdev 17:45
and Marc, do you think that’s because of what Carl said about the predictability for judges, and then is that perhaps working against those predictable divisions, because, of course, we see Munich being quite overutilized, and that leading now to delays in the, in the context of cases there
Marc Lauzeral 18:05
I think that there is a tricky thing about predictability, because I know that from the US, a lot of US colleagues told me that Europe is now kind of wild waste, you know, like we don’t understand really what is happening here. But regarding predictability, you know that, for instance, the Munich Local Division, as a president, was really extremely open about the fact that he was poor patentee. The only thing is, like, it was mentioned, it’s one year proceedings to get a first instance decision, less than a year to get an appeal decision. When you are looking at the overturning rate of the president of the Munich court, two decisions out of three are overturned. So, it means that now, if you want to get the most relevant decision at an earlier time, I would not recommend to go before the Munich local division, while when you are looking at the other local division except the German one, one decision out of three is overturned, so when you think about predictability, you’re going to say for maybe this judge I will have predictability, but when you are looking like the next two years, actually it will not give you predictability, because you’re going to say, oh, I’ve got a winning decision, but what will I do with that, because even the defendant knows that he’s going to have better chances of success to overturn this decision, so I’m not sure that that predictability is more like a German thing
Rajvinder Jagdev 19:30
but it’s sort of short-term predictability based on experience of the judges in national proceedings. We don’t have enough time in the UPC yet to be able to develop predictability within the court. There, did you want to come in, Car, or
Carl-Alexander Dinges 19:44
I would just like to make one statement. I don’t think it’s so.. I’m not quite sure whether this two on one rule is actually true for the merits of the case when it comes to the pattern infringement, but rather there have been, of course. It’s a bit of a clash of cultures, and that’s what one needs to understand, that there are so many judges coming from so many different jurisdictions that they look at the same question, and particularly on procedural questions from different angles and through different eyes, and that’s why I think at the beginning there have indeed been some decisions that have been overturned, but I think the predictability when it comes to the finding of patent infringement is still quite, at least in my personal opinion, it’s quite high. I don’t know the exact number on that, but
Rajvinder Jagdev 20:36
so you mentioned culture, that takes us deeply onto our next slide, language of proceedings. So one of the consequences of the German divisions being so popular early on is that at the start of the EPC, most actions were filed in Germany, and there are other reasons for that, around selection of languages and formal designation of English as a, as a language in certain courts, but we’ve seen that shift quite materially over the years, now over 24 and 25 Now the majority of cases, just over half, are conducted in English, and certainly the judges, the UPC, and we’re going to hear from from two of them later on, really promote English as the language, and that is to make the course accessible as possible to users, and Steve, this is a good opportunity to bring you back in on accessibility, because there is a cultural difference between UPC and its predominantly procedural or civil law approach to litigation versus what you’re used to in the US as a common law jurisdiction, and we’ll come on to other common law flavors in the UPC as well, but maybe you want to talk a bit about language first.
Steven Carlson 21:48
Yeah, I mean, the, there’s a lot of the litigants there are American, right? We are either being sued over there or bringing cases there, and you know, here we speak English, and patents are really funny documents with very intricate languages, we’ve kind of, we pushed the English language to the max, and when you construe claims, so much hinges on just these little nuances of language, and you want your panel to understand it, you want your panel to really understand English at a native language, at a native level, the courts have been sympathetic to requests to move the language to English if it’s filed in Germany. There’s a lot of motions to have it hurt in English. Those tend to win. There’s sympathy for the defendants being hauled in there, that they should be, have you know their own language being used a lot of eyes run Ireland, you know, Ireland’s not in yet, but I think if Ireland gets into the UPC, that could be a very hot jurisdiction, because you know you want to be able to relate in the English language to your judge who grew up with English, not as a second language, but as a native language, and so I think we’re all looking at Ireland as possibly being a very hot jurisdiction if they get the rat together, if they get online.
Rajvinder Jagdev 23:12
I think that’s a really good point about Ireland. Terry’s still in the room, but to answer her question about the UK joining the UPC, I mean, there are there are issues around the way the UPC agreement is drafted that are perhaps a barrier to that at the moment, but the second best option is Ireland, Ireland being a common law jurisdiction, and so it has a tradition of disclosure and expert evidence and discovery, and it’s important to note that the UPC rules actually do allow for some limited discovery, and so Rule 190 of rules of procedure allows for specific disclosure of documents, and so that’s an underutilized rule, and that’s partly cultural, certain courts and jurisdictions, because they don’t have a history or a tradition of discovery, don’t don’t really understand the utility, or don’t entertain it, but there was one case in the Hague Local Division where a claimant was able to get some limited discovery in relation to a an integer that they couldn’t make out without looking at the proprietary documents of the confidential documents of the of the defendant, and so they were successful in getting an order for discovery. They lost in the end, but the principal leases banks, and perhaps having a jurisdiction like Ireland, a division in Ireland that has a common law background, but might see more 190 applications being made and granted, and certainly in Germany, Carl, they are used for Sep Fran case, right, to facilitate disclosure of licenses in that context, and just on the English point, it’ll be great to hear from the three of you as to how. How you sort of feel about about listening patterns that are in English, but in French or German, you made a very good point, Steve, about about the detail of the language and nuances, particularly in the context of technical disclosure.
Carl-Alexander Dinges 25:18
Well, yes, if I may start, so interestingly, currently litigating patents, English written patents in English before the UPC courts is not as litigating before a native-speaking English judge, and it’s something that the judges make very clear themselves that they would like the parties to to lower their English level when they address specific questions, because all of those judges have have tradition of working in their specific native language, they have not, unlike many of the lawyers in the countries, have not dealt directly with international clients, and therefore had a part of their work being originally made in English, although the skills are very high, I would say we are at a very good level, and the judges do speak very, very good English, and things are very, go very well, but they address this question that, for example, bringing in a native-speaking witness can be some kind of difficulty, because, because if one, you know, if there are nuances in what the witness is saying, which are not understood by the audience, that’s that’s something I would consider, and rather opt for someone also being a second language speaker, and I must say I would, I would have an issue with litigating a French patent, and I would always try to bring someone in from from the from the original language who is with co-counseling or even first sharing the case, because it depends on the specific language skills you have yourself. I wouldn’t be comfortable with doing that for French patents.
Marc Lauzeral 27:34
Yeah, I think there’s going to be in French, from what I’ve witnessed, the only case you, you have a case in French, that the patent is in French, but usually the court are really, really open to change language of proceedings. If the defendant is an English-speaking party or a UPC representative are English-speaking, they are really open to do that. I think that all judges, some have been in private practice, even so. I would say that they speak very, they speak very well English, to be honest with you. I know that, for instance, Judge Camille Linear in Paris, you’re gonna hear this afternoon, and with Judge Grabinsky, both will address the question of UK and Ireland. So, I just want to make a little teaser, but Judge Linear lived in DC for two years, so they understand, of course, it’s not up to what a UK judge will, will, will, will understand the same thing that the UK judge will understand with the barrister, but, but I think the language barrier is not that difficult, to be honest with you.
Rajvinder Jagdev 28:35
I think conversely, it’s slightly harder for a native English speaker, because it is very much European English that’s used, and so using our idioms and colloquialisms often gets lost in in oral submissions, and so there’s a bit of adjustment for native English speakers to to actually properly advocate and get their points across to a panel of non-native English speakers, and again, that’s where a panel of Irish judges may change things somewhat, but still, by and large, most of the judges in the pool will be from other jurisdictions, and so adjusting the style of advocacy is important in the UPC. It’s not dissimilar to the EPO, I expect.
Carl-Alexander Dinges 29:22
but I must also say there has been a longstanding tradition in Europe to bring all bring different languages together, and then speak in French and French and German and English, because of the EPO, because in the EPO there have always been procedures with panels from many different countries, and they always had – they were obliged to speak those three languages, which are spoken at the EPO, and I think it turned out quite good. So, although the English is not as you would expect it in a patent. A trial in the US, it is perfectly good to litigate in the UPC in English, although there is no native-speaking country English-speaking
Rajvinder Jagdev 30:10
at the moment. So, while we’re talking about trials, you, you highlighted, I think, Mark, that it’s a one day proceeding and and so it is very different in style to what we’re used to in the UK, where a typical pattern travel will be four to five days, or in the US, where it’ll be even longer. Do you want to talk a little bit about the the nature of the proceedings and how they conducted,
Marc Lauzeral 30:39
so you have a written part and you have an oral part. Usually the judges are making some kind of pre-deliberation before the trial hearing. You’re going to have an entering conference when you, the procedural judge, the judge reporter, as we call it, will say, “Okay, I want you to advocate on this point, this point, this point. Then you’re gonna have a one day trial hearing, and it’s gonna be really, I must say, from my experience, you can discuss pattern for like four or five hours long, just purely with UPC representative making one hour advocacy, then it’s question and answer, so you’re gonna have what is good with this type of question, and then thing is you gain more predictability in a way that you can address some issue sometime in before national court, you are making two hours advocacy, then you get a decision, and you say, okay, I mean the judge clearly thought about something we never thought of, so it’s a bit surprising here, you know, where kind of you have the feeling where the things will go, and and you have like, like, like, like, it was mentioned sometimes the possibility, but it’s extremely limited to hear some, some expert, either through from the Nordic Baltic Division UK style, but it’s only limited to 30 minutes from my understanding, or like what you get in unique tube of experts when you, you, you are making just the expert discussing between each other, so yes, so it’s just one day intense trade hearing when you have to be really straightforward in your argument.
Rajvinder Jagdev 32:10
So the slides up at the moment, it’s nationality of parties, and Steve, you already mentioned that the US parties are the second largest group in the upc already, both on the claimant side and on the defendant side, and so I think it just highlights that the US is already in the upc, whether whether the parties like to the not on the defendant side, probably that they’re not so much of a fan of being in it, but there are also quite a few actions, infringement actions commenced by by by by us litigants, and I think we have a panel coming up later of in-house people who have some UPC experience, and I think they’ll be able to talk a little bit more about their experience of being in the UPC and the various divisions and the way in which hearings and the procedure generally is conducted. So one of the great things about the UPC is the availability of preliminary relief, and so you can get preliminary injunctions and other preliminary measures like seizure orders, and, and that is unique compared to the US, as we heard on our, on our earlier panel, and the grant rates are quite interesting, so based on our analysis, around a third of all, or just over a third of all, PI applications are granted, and then a further third, or just over, or just under granted in part, and so that’s a pretty high hit rate. Now, this isn’t just for preliminary injunctions, it also is for procedure orders, and that’s a tool that’s often used on the continent to gather evidence before filing a substantive claim. Yasmine, do you want to talk a little bit about this?
Yasmine Azzaoui 34:09
Yeah, of course. Yeah, it’s, oh, it’s really interesting to see that first, the most, like most of the PI cases are concentrated before the Dusseldorf and Hamburg and many local divisions today, in terms of volume, PI cases represent approximately 20% of all the UPC cases filed, and the average time to close a PI case is quite quick. It’s approximately 96 days before the Court of Appeal, the picture is quite balanced. So, today there are about 50% of the first instance BI decisions that are overturned by the Court of Appeal, and 50% that are confirmed, but this number will certainly be reduced with the time, because the Court of Appeal of the UPC is building its. Case law, and when we see the statistics of the reasons for refusal of granting a PI, most of the reasons are based on an insufficient degree of certainty of the validity of the patent or of the existence of an imminent infringement, and we can also notice that there are common procedural approaches between the local divisions. The local divisions are trying to limit the number of arguments that are brought by the applicants, for example, Do Munich Local Division has already said that the applicant must only bring like three arguments maximum. The Paris Local Division has also required that the applicant summarize their submissions before before the Paris Local Division, and we also see some minor differences in interpretation before the various local divisions. This is essentially concerning the requirement that the applicant must demonstrate that he has acted diligently after being aware of the existence of the infringement, so for example, the Dusord of Local Division stated that only one month is sufficient after being aware of the of the of the existence of the imminent infringement, whereas the mainland local division expressly departed from this from this and considered that two months can be sufficient to file a PI action after being aware of the of the infringement, so there are some minor differences, but like today the most of the of the PI cases are still brought before the German local division, and it will be great to fight to file more PI actions before the other local division, so as to build like more case law and to have more experience.
Rajvinder Jagdev 37:08
Okay, we have a question.
Audience 37:09
So, so Raj, just to clarify, what’s the x axis? So it says, okay, 1/3 PIs are granted, that’s significant.
Rajvinder Jagdev 37:20
Yes,
Audience 37:20
but
Rajvinder Jagdev 37:21
number of number of decisions
Audience 37:22
so there’s only been eight requests for PIs?
Rajvinder Jagdev 37:26
eight requests in Hamburg, and so you have to add them up, so it’s by division, and so it scales up. Do you have a
Audience 37:34
but the next, as you do the next two slides, the numbers just seem, are the cases just not, it’s too soon for them to have decisions, because there’s, we saw that earlier slide where they’re filing like 200 cases last. So they’re just too soon in the process?
Marc Lauzeral 37:50
that is what I’m mentioning earlier, given the folk screen of number of proceedings, when actually, when you are discussing cases, there are some you get more and more predictability now. You figure where it goes, but it’s not like you know 1000 cases, that’s that is, I think, pretty low quant what you just mentioned here.
Rajvinder Jagdev 38:12
So, what we’ve done in these slides, in the next three slides, we have collapsed down into the number of unique disputes, and it may be, for example, if you have three defendants in a particular action and you file a PI, well, that will count three times for each of the claimants. If you have two claimants in that, it’ll be six times in the official UPC count, but really it’s one dispute between two parties or related parties, and so what we’ve done here is collapsed down to two unique disputes rather than just a number of pleadings filed
Audience 38:51
three four and five, we had numbers,
Rajvinder Jagdev 38:55
yes, those
Audience 38:59
are motion kind of decision (inaudible)
Rajvinder Jagdev 39:03
Not unique cases, that’s right. So the UPC report slide, so this, and then the blue background ones are the number of actual pleadings filed. Essentially, that’s the
Audience 39:18
we’ve been in business for a long time and there’s a difference between a motion and case.
Rajvinder Jagdev 39:25
There is a whole panel discussion we had about the IC of the UPC and the deficiencies with it, that things are improving. Docket Navigator does a good job, for example. It ought not to be that hard. It ought not to be that hard, but it’s proven difficult for the guy. So, one of the, one of the big changes this year to the EPC is actually a move over to a new system built by the EPO, so things are better, and we’re hoping that we’ll get better quality data, as you do for EPO. Seedings out of the new system, so the next time we have this this UPC day, we also have some, some better, more accurate data that’s aligned with the UPC official statistics, but for now we leave it to the to the market to come up with the data. There are various source
Wayne Stacy 40:18
So, Raj, I wanted to ask a question, just kind of put things in context. How long duration of time from filing to an enforceable preliminary injunction?
Rajvinder Jagdev 40:32
That is a very good question. So, before I answer that, it’s helpful to set the time frame for the written procedure for a proceeding on merits, so the written procedure takes eight months from start to finish. For a preliminary injunction, collapses right down to around two months, two to three months, which is incredibly condensed, and so that is the application for the injunction. And then the defendant will have an opportunity to make non-fringement arguments, and also raise any invalidity arguments, that’s quite until you have about four weeks, four to six weeks to pull that all together. So, there isn’t very much time at all to do that, and you’re probably going to be faced with a proceeding on the merits at about the same time, so they’re having to fight on two fronts. The system is very much weighed in favor of the patentee claimant at the moment, given the front-loaded nature of it. It puts an awful lot of pressure on defendants to act promptly and quickly and efficiently in responding to applications or for PIs or even substantive infringement claims,
Wayne Stacy 41:43
so the follow-up piece to that question is, you know, in the US, if you hear about preliminary injunctions, it’s like Bigfoot. I’m not sure that I’ve seen anything other than pictures of them in a decade, but they theoretically exist, but then they come with this enormous hammer of having to post a bond, which could be 10s or hundreds of millions of dollars. Is there anything equivalent that you have to post for the UPC?
Carl-Alexander Dinges 42:13
Yes, yes, there is. So, there is a large discretion for the court to post a bond for for preliminary injunctions in and I think it would depend on the industry, so in going back to what we know from the from the national patent litigation, in very many cases there is no bond in place as long as you believe that the the applicant is able to ultimately recover for damages which are which incur based on the injunction, and for example, in originator generics cases, there would not be a bond for the injunction and obviously, in an NPE case, request the NPA requesting for an injunction against a large company being active on the market, there would evidently be a huge bond recovering what what is at stake, but the requirement to actually get a bond is that you produce evidence on the damages, which are, which likely occur based on the preliminary junction, and that is something which companies usually don’t like, so it’s still, you really get into, you need to get into a discussion at what height should a bond be placed, but even if you don’t want to disclose everything, I think there’s ways to establish facts that there should be a high bond. Yeah.
Audience 44:03
Sorry, I have a follow-up question. So, if their bond or estimate of damages is large, that kind of suggests that the harm caused by the injunction is irreparable, and if there is irreparable harm being caused by the injunction, should that factor into the balancing of interests when considering the grant of a PI.
Carl-Alexander Dinges 44:28
I think, yes, it should, and I have.. I don’t really have the statistics on whether there is cases in which the injunction was based on this, this aspect, which you raise, not necessary. So, so ultimately, the question whether an injunction is issued is whether there will be irreparable harm for the claimant if it’s not issued, which is something that would be taken into account, and therefore in those situations. Where you don’t actually protect a certain market, and you cannot show there’s damage for you, it would be quite an obstacle to get a preliminary injunction.
Rajvinder Jagdev 45:18
And actually,
Carl-Alexander Dinges 45:19
do you have experience with that? Have you seen injunctions?
Rajvinder Jagdev 45:23
I don’t, but what I was going to say is that there is an effort amongst the judges to dissuade parties from seeking preliminary injunctions because of the burden they place on the court, but it doesn’t matter because you can get to a final injunction within 12 months and that takes us on to our our stats on on infringement, so the infringement outcomes in the UPC are also quite, quite interesting, so just over half of all infringement claims lead to finding a very valid and infringed patent. Carl, do you want to talk briefly about about this slide and the success or the yes for various divisions?
Carl-Alexander Dinges 46:13
So I think this is a significant finding that over half of the cases are found to be infringed and valid, the patterns to be infringed and valid, which significantly differs from from what we’ve experienced in national courts, although we had quite high rates of infringement findings in Germany earlier, we also had a high rate of invalidations of patents, and there was an injunction gap. Several cases have been settled, but I would say by large we didn’t have that high overall success rate. What is interesting also is that if you look at the different local divisions the tendency is roughly the same. I mean, of course, if you go if the case load is too low you cannot really take too much out of the numbers, but also in Justin of Munich, Manheim, and Paris, it’s roughly, it’s roughly about 50% of the cases being held infringed and on valid patterns.
Rajvinder Jagdev 47:38
Steve, very quick, I’m conscious of the time, how does this compare to the US?
Steven Carlson 47:43
Well, I know that for patents trying to survive the PTAB, at least as of 2024 now it’s vastly changed over the last few weeks, even. But it was about only about 1/3 of patent claims were surviving the PTAB, and then you got to survive the Discord and find it in French, so it was a very small percentage of people actually winning in the US, compared to this slide, which is, you know, over 50% of claimants were winning. So, you know, going back to Wayne’s point, faster, cheaper, better, stronger. I mean, you’re way better off from just running the odds in the UPC.
Rajvinder Jagdev 48:19
And to finish, we do have a slide on validity as well, Marc, If you want to talk briefly about this
Marc Lauzeral 48:28
I don’t want Richard to be angry at me, so I’m going to be super quick. Just to mention that maybe, compared to national code, the interesting thing here is a possibility to have a pattern valid as amended, and you can see that the figures are really high, and I think that the ability to be able to amend the patent directly before the court, which is usually not a possibility before national court, allows are strictly limited allows for your patent to survive the proceedings, and the fact that it’s the same proceeding usually that the infringement you will not lose, you know, you would not have any squeeze argument when you’re going to lose your right of infringement, it’s a good thing for the patent
Audience 49:01
Can you ammend the patent… (inaudible)
Marc Lauzeral 49:05
Yes, yes, completely, completely. And that’s the tricky part. When you have the patentee, because you have a huge frontloaded system, you are kind of limited in your argumentation, but if you have to defend several states of claims, and when you are thinking sometime about, like, for instance, in the life science industry, things when you have huge application with different diabetes indication, dosage regimen, everything you can have really upper hand when you’re the patentee
Carl-Alexander Dinges 49:32
but you get only one bite in the apple, so you can amend it at the stage of the reply to the statement of defense, and that’s it
Audience 49:43
If I may, a quick follow up on this. So, how does this work in a BSH long arm model? You know, you’ve got an infringement allegation for, say, Spain in your UPC case, and how do you get to a valid as amended patent when the Spanish court has exclusive jurisdiction over the validity of the patent.
Marc Lauzeral 50:04
I think you have to apply in any way Spanish law due to BSH decision. So, if the Spanish law does not authorize that, then you’re going to have to say there is no infringement, but that’s my top of my head reasoning.
Rajvinder Jagdev 50:17
So, I expect BSH will be covered on your panel, Ari. At the end of litigation strategy, it’s a new twist on European patent litigation with long-armed jurisdiction, and opens up some interesting possibilities. There any other questions regarding this?
Audience 50:46
(Inaudible)
Wayne Stacy 50:46
so can you repeat the question, so we get it on on tape. It’s such a good question.
Rajvinder Jagdev 50:52
The question, the question is, are patent is the patent amended in the UPC, in the tribunal, or back at the EPO, and the answer, at the tribunal
Audience 51:09
I’m a judge from IP High Court of Korea, so perhaps we have very similar patent system as your many European countries, so this amendment can also be appealed to the court, and the court decide on whether we affirm that amendment, so and the amendment is limited to fixing obvious errors or limiting down the scope of patents, and patent holder cannot add new things on the patents is how that works
Rajvinder Jagdev 51:44
I think that’s exactly right. Thank you. Yes, the amendments are narrowing in nature, only limiting the scope. Obviously, you can’t expand this or rewrite completely, though there have been cases, certainly in UK litigation, where we have managed to effectively rewrite the claim in in the course of litigation, very good. Well, thank you very much, everyone. If you join me in thanking the panel.