Perhaps it’s just me becoming more and more impatient with age, but I increasingly wonder why it is that so many things take so long in my home country. The litany is endless, from Berlin airport via Stuttgart 21, to any simple construction site on any German highway. From the formation of our government to passing even the most uncontroversial law. Not to mention the ever more badly needed switch to renewable energies, wherever possible. And, last but not least, nullity proceedings before the Federal Patent Court. The average pendency has meanwhile increased to almost 27 months (from an already deplorable 21 months in 2009). Even worse, in those areas where most money is at stake and where legal certainty is most desperately needed, i.e. in the fields of chemicals/pharmaceuticals and electrical engineering, it now takes no less than three years before a nullity action is decided.


Is Germany, more specifically, are the parties of nullity proceedings, is industry, are the patent attorneys, the Federal Government, the Federal Patent Court happy with the status quo? I believe not, but sometimes I find the public silence on this topic deafening. Or is it just me who fears that Germany’s good reputation as one of the best battlefields for patent litigation in Europe is going down the drain? Does anybody worry about the consequences for the German economy?

Yes, I have heard that honorable ladies and gentlemen from the Board of the Patentanwaltskammer (PAK, German Patent Bar Association) have more or less privately and certainly most politely expressed their concerns about the current state of affairs in January. I say “more or less privately”, since nothing whatsoever has been published about the results of this meeting in the monthly Journal of the PAK or otherwise been publicly communicated. The PAK seems to place all of its bets on backroom diplomacy, yet the devastating lack of success of such initiatives over the past ten years suggests to me that this is simply not enough. I am also missing a public outcry by industry (everybody, please forgive me if I have overlooked something, I would only be too happy to be corrected and will readily publish any of your initiatives on this blog).

The judicial consequences of this bad administration of justice cannot be overstated. Imagine yourself being a company who wants to bring a new product on the German market. Suppose a patent of at least doubtful validity stands in your way. What can you do? Your nullity action takes almost three years, not including appeal. And if you dare enter the market before you have cleared the path, you are at the mercy of the infringement courts, of which at least some praise themselves as having a “very restricted” practice of staying their cases pending the outcome of your nullity action. You run a high risk of being injuncted and pushed out of the market before the patent is revoked.

Will your response be: “Well, well, well, but is this such a big problem? Each patent has at least been thoroughly examined by either the German or the European Patent Office. So don’t these well-examined rights deserve being respected?” – Hmmm. The validity of this argument obviously depends on the quality of the patents granted by the respective offices. And here lies another serious problem. The EPO, for example, has almost doubled the number of granted patents over the past 5 years. According to the EPO’s Annual Report 2018, 66712 patents were granted in 2013. In 2018, the number of patent was 127625! Conversely, the number of examiners has almost exactly stayed the same (4221 in 2013, 4276 in 2018). This means that compared to 2013, each examiner now has half the amount of time available to examine and judge each application. And given that it is easier to comply with applicant’s wishes than to write a decision of refusal, you can guess what this trend means for the average quality of the search and examination process.

But even if the quality of the granted patents had miraculously stayed the same for the last 5 years, this does not mean that examination by the EPO or GPTO makes a patent fire-proof. On the contrary, the percentage of total or partial invalidations by the Federal Patent Court is significant.

How significant is it? This takes us to the results of a small inquiry (“Kleine Anfrage”) of German MPs in the Bundestag, to which the German Ministry of Justice and Consumer Protection (BMJV) responded with the following statistics, as reproduced in the most recent edition of the journal of the PAK. The second column of this statistic refers to the total number of cases disposed of, the third column to complete revocations (or more pedantically, “declarations to be null and void”), the fourth column lists the percentage of revocations to total number of disposed cases, and the fifth and sixth column do the same with partial revocations.

The impression you might get from these statistics is that only a relatively small (in 2018, 24% + 16%) proportion of patents is wholly or partially revoked, whereas the rest is maintained. But this is not so. In fact, the BMJV’s statistic is – apologies for being so outspoken – quite misleading. This is because most cases before the Federal Patent Court are not “disposed” by a contentious judgment, but by withdrawal or some other sort of settlement. In particular, patentees may wish to settle a nullity action – e.g. by granting a free license – when they know their patent is of doubtful validity. Thus, while the number of “disposed cases” is somewhere from 200-250, the number of actual decisions per year is only in the order of 100. For example, in 2015 there were 93 judgments, of which 47 (50%) ended with total revocation, and 32 (34%) with partial revocation. The patent was maintained as granted only in 17 (18%) of all cases!

This trend has not significantly changed over the last 5 years. One simply cannot assert that even the simple majority of patents that are seriously attacked before the FPC will withstand nullity plaintiff’s challenges. True, only a small portion of patents is enforced/attacked by a nullity action, but there are voices, particularly in academia, who take the view that these invalidation rates will not be significantly different for the “average patent” granted by the EPO/GPTO. Their argument is that a nullity action is typically the by-product of an infringement action initiated by patentee, and the patentee will only select a patent for enforcement which it considers at least prima facie valid. I am not sure whether this applies to all or even to most cases, but such arguments should not be easily brushed away.

What could and should be done?

In my view, several things could and should be done, both on a legislative and administrative level as well as, possibly, inside the Federal Patent Court itself. Only if everybody concerned works hard to improve on the current deplorable state of affairs, can progress realistically be expected.

On a legislative level, I wonder if something can be done to relieve the judges from part of their work. For example, nullity actions are presently decided by the Federal Patent Court in a composition of five judges, including three technical judges. Ordinary technical appeals are decided by a panel of four members. In contrast, the Boards of Appeal of the European Patent Office do about the same job in a composition of three members. Why not reduce the nullity boards (and the technical boards of appeal in the FPC) to three judges? I do not think that this would significantly affect the quality of the court’s decisions, but it would free precious time of the judges for studying other cases, decision drafting etc.

On the justice administration level, it seems both inevitable and very urgent to me that more technical judges be appointed, particularly in those fields where the demand is highest and the cases are most complex. Appallingly, the exact opposite has happened in the last couple of years. The two chemical appeal boards have been fused into one and retired judges have not been replaced. Once again, we cannot continue like this!

Finally, the Federal Patent Court should perhaps also consider which internal measures could be taken to increase its output. According to its Annual Report 2017, there are currently 58 technical judges (including presiding judges) at the FPC. Assuming that about 100 decisions in nullity cases are currently drafted per year (i.e. about 2 per reporting judge on the average), it might not be unreasonable to aim for a (voluntary) increase to three or four nullity decisions per year and judge. I am perfectly aware that the judges also have other duties, e.g. writing decisions in appeal cases, voting, drafting the mandatory preliminary opinion etc., and that many cases are extremely complex and take a lot of time to prepare, but I would still argue that if every technical judge manages to write about three nullity decisions every year as a rapporteur, this would significantly reduce the current backlog.

In summary, I would think that it is ample time for another “Ruck” (jolt) to go through Germany.

– The legislator should consider reducing the requisite number of judges per decision to three. Less time to spend in hearings means more time for decision writing.

– The Ministry of Justice should finally and urgently recruit more technical judges. It will pay out in more than one respect: the existing judges will be more motivated and more cases will come to the FPC in the long run, which will help to increase the court’s revenues.

– The judges can also contribute. It is true and correct that judges must be independent, and nothing that I have suggested here should be taken as “pushing” them in any way. I just think that our constitutionally guaranteed rule of law also implies a general duty to guarantee justice to everybody (allgemeine Justizgewährleistungspflicht). I am confident that judges can and will contribute their part to fulfill this duty.

– And, finally, the parties – and by that, I mean primarily myself and my colleagues – should also help in facilitating the judges’ work by writing fewer pages and reduce the number of auxiliary requests to a sensible minimum. Being patent attorneys, we should not forget that we too are in the first place an independent organ of the administration of justice (Sec. 1 Patent Attorney Act).

If everybody contributes his fair share without waiting for others to move first, we have a real chance to get rid of the existing backlog of cases, ensure proper and quick justice to the parties in need thereof and bring our patent dispute resolution system back into a leading position in Europe. Auf gehts! / Let’s go! / En marche!


To make sure you do not miss out on regular updates from the Kluwer Patent Blog, please subscribe here.

Kluwer IP Law


  1. Why such modest demands?
    EPO Boards of Appeal manage around 10 decisions per Board member each year.

  2. In his list of things that take too long in Germany, the author forgot to mention rulings by the Bundesverfassungsgericht…

  3. Putting aside the exceptional cases where emergency interlocutory (interim) injunctive relief is appropriate, Thorsten, was there ever a period in Germany when the infringement courts had the benefit of a decision on validity BEFORE they enjoined the accused infringer? You see, I’m thinking that in England it is the very opposite. Was there EVER a period in England when a judge would enjoin an infringer before the validity issues had been explored at trial after discovery and cross-examination?

    I’m thinking that it is a cultural thing, associated with the notion (in Germany) of exclusive rights that have aleady been “examined” rigorously within the Patent Office and so are prima facie valid. Given this presumption, what does it matter (think the old white men in charge of the patent litigation system in Germany) if the judicial exploration of validity takes ten years or so? Indeed, is it not “Good so” (because the parties can budget proportionately, and find a way to settle the dispute during its pendency).

    In England, patent litigation is different, and set by history. Up to 1978, under the patent law of 1949, the Patent Office was required to examine rigorously for novelty and clarity but to leave obviousness and enablement to the courts. Why, because such issues were thought to be judiciable only in full-blown inter partes proceedings, assisted by discovery and cross-examination). Unfair to inventors, it would be, for a Patent Office Examiner to refuse an application for obviousness. Unfair to the public, to endorse as presumptively valid, an application checked for obviousness only by an examiner sitting in an ivory tower.

    This history explains why courts in England first come to a final (appeal is typically pointless) decision on validity, after discovery and cross-examination at trial, within 15 months of the start of the dispute, and only then enjjoin the infringer. For that short sharp dispute resolution process though, you certainly need a deep pocket.

    As Hugh Laddie reminded us, with is eyes very much on Germany, do not make the mistake of supposing that, in England, the courts are “a branch of social services”. Perhaps there is great need for “a branch of social services” when there is a patent dispute?

    Two different jurisdictions. Two different philosophies. As the saying goes “You pays your money and you takes your choice”. This is Europe. Vive le difference. Good so. Horses for courses, I say

  4. The Federal Constitutional Court’s list of “decisions envisaged for this year” is not without a reason nicked “Lügenliste” by the members of the German Rechtsanwaltskammern (German Regional Attorney’s Bar Association).

    And that goes without saying, but the UPC/UPA-complaint before the BVerfG has no impact on legal certainty for cases under dispute, and a reference to that would just divert attention away from the main issue, being that most European states have neglected investments into their legal systems.

  5. On a more controversial note, how about abandoning bifurcation in favour of a more UK-style system: where infringement and validity are considered together but there is a possibility of obtaining a preliminary injunction ahead of a full trial?

    As I read the statistics presented above, bifurcation is systematically denying timely access to justice for those parties who merely wish to “clear the way”. So why not ditch it completely?

    And why on earth set up the UPC so that it can also develop a fast track for infringement and a slow lane for validity proceedings? (Though it has to be said that, in my view, that is probably the least of the myriad of problems with the UPC … and so perhaps it is no bad thing if the BVerfG waits until the dust settles on Brexit, so that the facts surrounding the legality of the UPC system are known.)

  6. When I worked as a “Kandidat” in two senates at the FPC, I was impressed how rigorously the judges have studied the cases, even if it is the 15th appeal by a private person who is a serial inventor of perpetual motion machines. I find their rulings much more balanced and nuanced than what the EPO’s BOAs issue on average. That said, I am not sure whether we really need such an rigorous analysis of each case. As some scholars put it, one should have the right to a fair trial, but not to the perfect trial.

    The other issue is that you cannot win an election based on a patent-related topic. Hardly anyone outside our community understands the issues or what a patent is to begin with. The general public simply does not care about it enough for mainstream politics to pick it up.

  7. “And given that it is easier to comply with applicant’s wishes than to write a decision of refusal, you can guess what this trend means for the average quality of the search and examination process.”

    We should also not forget that since April 2018 (and with retroactive (!) effect as from 01 January 2018), the production points awarded to EPO examiners for refusal decisions has been reduced from previously 2 points to now only 1 point.

    The previous counting of 2 points per refusal decision was done in recognition of the fact that a refusal decision requires significantly more work (typically a number of Art. 94(3) EPC communications, then summons, holding of oral proceedings, a lengthy decision on possibly multiple auxiliary requests to be written) than an intention to grant.

    Hence, since this difference in workload between a proposal to grant and a refusal decision is no longer recognised, there is now absolutely no reason for a sane EPO examiner to burden herself or himself with the hassle of a refusal.

    The reason given by senior EPO management (COO HBC) as to why refusals were now recognised identically to intentions to grant, she said that this was to harmonise award point counting with production counting.
    When asked whether harmonisation was the relevant factor to consider, when what was at stake was a difference in workload, a silent smile but no answer was given.
    When then asked whether the reduced recognition of refusal decisions meant that less effort was now expected to be put into these decisions (such as by providing less reasoning, or no reasoning at all), she refused to answer the question.
    Now there is some surprise at management level that a recent staff survey at the EPO has revealed sub-zero confidence levels of the examiners in their senior management…

Comments are closed.