Call to readers: let's talk about patent quality

Last week, the Intellectual Property subcommittee to the U.S. Senate Committee on the Judiciary held a hearing to inquiry how the U.S. Congress can prevent the issuance of poor quality patents [video recording and testimony transcripts here]. The hearing was part of a series to educate the subcommittee on the current state of U.S. patent law in light of the proposed STRONGER Patents Act of 2019.

Several interesting comments and suggestions for improvement were made during the hearing, which are briefly summarized below. The hearing is also a good opportunity to reflect on the quality of European patents, and this Kat would like to invite readers with experience or expertise to share their thoughts on this subject. The call is explained in further detail in the second part of the post.

The importance of quality is universally recognized (Meme credit:

The hearing before the subcommittee

Testimony was first given by Mr. Drew Hirschfeld, Commissioner for Patents at the United States Patent and Trademark Office (USPTO). Mr Hirschfeld discussed various measures the USPTO had taken or is planning to take to increase patent quality, such as providing additional resources to examiners, issuing clear guidance on patentability and exploring artificial intelligence capabilities.

Next, testimony was given by Professor R. Polk Wagner, Professor Melissa Feeney Wasserman, Ms. Teresa Stanek Rea and Professor Colleen V. Chien. The witnesses suggested various ways in which patent quality could be improved, such as:

  • Setting a clear and uniform standard for claim construction [Wagner testimony, 3]
  • Changing the USPTO's fee structure to create less reliance upon post-grant fees [Wasserman testimony, 17]
  • Increasing the time allotted to examiners to review applications [Wasserman testimony, 22]
  • Facilitate collaboration between examiners, especially from counterpart patent offices [Rea testimony, 4]
  • Relying on pilots to test different solutions [Chien testimony, 1]
  • Create the option for applicants to indicate that the patent will only be held for defensive purposes in exchange for a discount in maintenance fees [Chien testimony, 2]

After the testimonies, the senators posed questions to the witnesses. Professor Wagner explained his view that Congress should be careful about instituting sweeping reforms because they will affect different industries differently and changes might well work to the detriment of smaller companies and individual inventors [video recording at 01:03:00, referencing earlier work in which he surveyed how the change to a first-to-file system had affected individual inventors, here].

Senator Blumenthal expressed concern about perceived abuse of the patent system by pharmaceutical companies. He asked the panel (i) if there was agreement among them that pharmaceutical companies sometimes use patents in an anti-competitive manner and (ii) whether this is an issue of patent quality [video recording at 01:12:15]. Several panel members seemed to agree that there is indeed potential for "strategic behaviour" by pharmaceutical companies, but it was pointed out that more than anything such behaviour arises out of the interaction between the patent system and the regulatory framework (or lack thereof). Professor Wasserman made the interesting suggestion to let the patent office spend more time on applications that are likely to be listed in the Orange Book, i.e. the list of drugs approved for marketing [video recording at 01:16:30].

Professor Chien compared practices and quality between the USPTO and the European Patent Office (EPO) [video recording at 01:07:30, drawing on a 2018 paper, here]. According to Professor Chien, the key difference is the early availability, before the EPO, of a search report that allows applicants to make decisions on continuation of the application at a very early stage of the process. This and other differences, she says, lead to high satisfaction with the EPO:
The surveys I've seen […] have shown that the EP [European Patent] is of the highest quality and also of the highest satisfaction […] There is a sense of a respect for EP process.
That's a great compliment to the EPO and the hard work they are putting in to make the European patent system as strong as it can be!

What about Europe?

The EPO has reported very high (80+%) satisfaction rates for some years now [here]. And it has been ranked first for patent quality among the world's largest patent offices for seven consecutive times in industry surveys [here; and see a 2011 survey carried out for the European Commission that found that "Companies assigned the European patent system the highest overall rating" (at 60), here]. That is good news for Europe, because the importance of patent quality is one of the few things that all patent professionals seem to agree on.

Still, there is data that suggests a less rosy picture. In a recent paper, Professors Henkel and Zischka find that between 2010 and 2012, the German Federal Patent Court fully invalidated 45% of patents challenged before it and partially upheld another 33% [here]. Based on survey evidence, they estimate that these chances of invalidation can be extrapolated to any 'randomly picked patent'. Looking at Europe-wide data in the period 2000-2008, Cremers et al also found fairly high invalidation numbers, especially in the UK, though nowhere near as high as Henkel and Zischka [here]. Could it be that European patents are getting more likely to be invalidated in litigation?
Infringement and invalidation rates (2000-2008) according to Cremers et al (Source: Springer)

This Kat has the impression that the answer might be yes, though that is largely based on anecdotal evidence. And it seems to be contradicted by consistent survey findings—though it is possible, if a little counter-intuitive, that European patents are perceived as being of high quality despite (apparently) being invalidated approximately half of the time.

That is why he turns to you, dear reader, for help. Readers familiar with the European patent system are warmly invited to share (i) empirical work on the quality of European patents, (ii) any personal experiences and observations, or (iii) any other material you think is relevant to European patent quality. You can do so by responding in the comments below or, even better, via e-mail [contact details here]. Naturally, all submissions will be treated confidentially.

Your input would be greatly appreciated! Insights gained from your submissions will be the subject of a follow-up post.
Call to readers: let's talk about patent quality Call to readers: let's talk about patent quality Reviewed by Léon Dijkman on Friday, November 08, 2019 Rating: 5


  1. The most compelling evidence is the EPO's own quality report from 2018. Once you have delved through the extensive timeliness and user satisfaction data you find the following:

    Figure 36: Percentage of patent grants found compliant by quality audits (page 41)
    2016 - 85.4%
    2017 - 84.7%
    2018 - 76.6%

    This shows that in 2018 there was a startling drop in the quality of patents granted by the EPO. According to the EPO's own data almost one in four patents that it granted in 2018 were not of suitable quality. That is astounding. The EPO seems almost to gloss over this data despite it being, to my mind, the most important aspect of patent examination. It prefers to focus on customer satisfaction and timeliness as measures of quality.$File/quality_report_2018_en.pdf

  2. Bad patents that are wrongly granted always attract the most attention. However, to get a rounded view of the quality of examination, you also need to consider good patent applications that are wrongly refused. I find this is a particular problem in US prosecution, where some intransigent examiners seem to see it as their role to come back with successively less plausible objections until the applicant runs out of patience or money.

    One way to iron out individual inconsistencies would be to compare rates of grant / refusal / number of examination reports between different examiners working in the same technical field, if that is not already done.

  3. What is "quality" depends on who you ask. Consider, for example, the "quality" of management, within the EPO, of amendments made after filing of a patent application.

    An Opponent would say that the EPO is right to be ultra-strict. An Applicant would say that in being so strict, the EPO is wrong.

    So what about a "neutral" observer, say, a judge who hears consolidated infringement/validity cases. That judge would presumably say that the EPO is at an optimal level of quality under Art 83, 84 and 123(2) when it accurately mediates between "fair" protection for Applicant and "reasonable" certainty for everybody else.

    And EPO Quality Managers? Presumably they decree that no patent ever issued by the EPO should ever be revoked by a court, post-issue, for added matter. Which presumably dictates, in consequence, a FAILURE of quality standards at the EPO, through the loss of any balance between fair protection and reasonable legal certainty.

    Or, to put it another way, how can it happen that the established case law of the EPO requires Applicant to prove a negative, and to a "beyond any doubt" level of proof, namely that the requested prosecution amendment does NOT add matter. What sort of "quality" does that case law force?

  4. There is no doubt that EPO Examiners are presently under more productivity pressure than 10 years ago, and that has led to increased impatience with applicants, whilst at the same time Boards of appeal are less likely to reconsider the entire case de novo to give the applicant a fair second opportunity. In my experience Examiners are less helpful in explaining what scope of claim could be allowed, and they often adopt quite a negative mindset which is difficult to change. Sometimes it is clear that EPO Examiners are learning the sorts of tricks that opponents use in formulating very imaginative objections. Sometimes I think Examiners see it as some sort of argument, especially when they start refusing to allow amendments or raise entirely new unsearched matter objections for new amendments. That did not happen 10 years ago where things were much more friendly and cooperative, and Examiners did not use their discretion in such a forceful way. I would therefore say that the EPO has become a less friendly and more legalistic. In my experience it rarely grants claims which are too broad, but it does struggle with complicated cases which would require more depth of understanding and necessarily more Examiner time. However I suspect Examiners just don't have that extra time to give, and so in this respect standards have slipped.

  5. Quality, what do you expectFriday 8 November 2019 at 18:45:00 GMT

    It would be interesting to see how many patents are revoked or severely limited in oppositions before the EPO on the basis of documents which were not found during the initial search. No need to wait to see what happens in national courts. The latency time of those results is much too long.

    That some examiners have after 10+ years never or very rarely refused an application does not show that quality is their prime concern. It is easier to grant by closing both eyes than to write a refusal. The points are all what matters!

    When the former president boasts about the fact that under his tenure 82% more patents have been granted, and that the present one wants an increase of production of 20% any discussion about quality is quite rhetoric.

    Given the time/action which is allowed to examiners, you cannot expect wonders. The new recruitment policy, and the departure at a more and more young age of experienced examiners will certainly not improve quality. But the rules of modern management are merciless, and the pseudo managers of the EPO are proud of their ideas.

    The problem is that management dwells in so-called positive user satisfaction surveys, whereas the audit figures tell a different story, see the quality report above. Simply dwelling on past glory is not enough. If in the past, the EPO was renowned for its quality, present management is satisfied if the quality is considered better than that of the USPTO.

    And in order to gain support from the AC, the present management gives a quite apocalyptic picture of the financial situation! That the premises underlying the new financial study are anything but reasoned or reasonable is a matter of fact. What do you expect?

  6. 1. In my general experience, both US PTO and EPO examiners are competent, motivated, and helpful.
    But the EPO system has a major advantage over the US system when an applicant encounters an intransigent examiner. If the US examiner is a primary examiner, there is essentially no review of that one person's action short of appeal unless that action is so outrageous that management can be persuaded to intervene. In the EPO, examination is by a three-person panel, so that even if the lead examiner is difficult, his/her colleagues may well keep extremes in check.
    Also, it's possible to compare allowance rates for US examiners, through websites such as BigPatentData (though that is a subscription site) or the now no-longer-updated Examiner Ninja. If you look at a particular art unit (examining area), you may well see allowance rates varying by a factor of two between examiners. This should not be the case - these examiners are, more or less by definition, handling highly similar applications, randomly assigned, and the allowance rates should be similar. The US PTO has been very poor in dealing with this. The EPO examining division system seems to avoid this again by the use of three-examiner panels.
    I'd like to see the US PTO make a concerted effort to review examiner statistics and try to "level the playing field" for applicants, so that your chances of allowance are not so dependent on the examiner you get. Three-person panels are almost certainly a non-starter, though.

    2. One should use caution when comparing invalidation statistics between the US and Europe (or essentially anywhere else).
    In the US courts, patents are statutorily presumed valid and invalidation requires proof of invalidity by the "clear and convincing" standard. CAVEAT: the Patent Trial and Appeal Board in inter partes reexaminations applies the looser preponderance of the evidence standard.
    But, as I understand it, almost everywhere else, there is no presumption of validity, and so a patent can be more easily invalidated.
    So a higher invalidation rate in Germany, say, than in the US does not to me necessarily imply that EPO patents are getting weaker.

  7. Quality, what do you expectSaturday 9 November 2019 at 09:22:00 GMT

    Dear Derek,

    I do not want to take your illusions away, but the notional three men examining division only exists on the paper. Once the first examiner has signed, and there was no oral proceedings, there is little chance that the second member or the chairman will say anything against the decision taken by the primary examiner. If they would do so, they could be faced by retaliation by the primary examiner in their own cases. It is a sort of equilibrium between the various interests, the prime interest of each and every examiner being that the production targets are achieved.

    Under ISO 2000 there has been a procedure called “CASE” in which divergences between the members of the division have to be recorded. There is no surprise that the result of CASE is that nearly 100% of the files are considered showing no defect whatsoever. Some directors/team managers have asked not to reach the level of 100% and to record some divergences, as it is not plausible that there are barely any problems, when the audit shows that conformity has reached a low level of ¾ of the “products”.

    The new VP1 is pushing the introduction of a “Collaborative Quality Improvement” scheme, with the aim to force the divisions to discuss and record the discussions between the members of the divisions before granting a patent. As such the aim is to be fostered, but the downside is that the time and efforts spent on those discussions will not be rewarded at all. At the EPO the pilots work always very fine, but after roll-out, reality hits and the results are far from the expectations. This was the case with CASE (sorry for the pun). The “Collaborative Quality Improvement” scheme will be rolled out office wide. The result is foreseeable like for CASE.

    In case of oppositions and of oral proceedings in examination, there is a better chance of collaboration between the members of the division. But there is no guarantee. In principle, all members see the annex to the summons, but it is not infrequent, that on the day of the oral proceedings the division has, after careful examination, a different opinion. This is not bad as such, and to be encouraged, but the work should not be done on the day of the oral proceeding, but well in advance. Reward for this preliminary, but important work, nil! This way of doing is legal, but goes to the detriment of the parties.

    In the early days of the EPO, the aim was “Applicant friendliness”. With the present production pressure, this aim has become obsolete. What matters now is “Production friendliness” so that the nth plan of the upper management is fulfilled so that they obtain the bonus they think they deserve.

  8. In response to Quality, what do you expect, the fact that some Examiners never refuse a case, does not mean that quality has suffered. It is the best Examiners that offer claims of very narrow scope which satisfy everyone. Third parties are not unduly hindered, and the applicant has some protection. Such wise Examiners are now rare, who understood that since most EP patents are not opposed or litigated, no third parties are interested or really care. That means the applicant might not need a huge scope of protection, and protection granted will not inconvenience anyone. Granting patents of fairly narrow scope will satisfy everyone is most instances and the system keeps going without too much fuss

  9. I beg to disagree with Sing-A-Pore. I have always thought that the job of a representative is to obtain the largest protection possible for his client, and that is what they are trained for when sitting the EQE. If depending on the state of the art, only limited protection is available, then why not. But certainly not in general.

    An examiner never knows whether the application he is deciding upon will be commercially interesting or not. Simply banking on the absence of an opposition or litigation is not what an examiner should have in mind. This is for me not a good or wise examiner, but an examiner wanting to get its points as quickly as possible.

    How can the examiner decide from his desk that the patent he is granting will not inconvenience anyone? This seems quite preposterous to me, not to say arrogant.

    I would go as far as to claim that the representative and the examiner are dragging money out of the client’s pockets by granting a patent which has hardly any value. This is not the job of an examiner or of a representative.

    I maintain that an examiner having never refused an application is not a good examiner. It is clear that writing a refusal which withstands scrutiny of a board of appeal is a task which needs more time than to decide a grant for reasons which are never made public. Hence my comment on the quick points.

    And when you know for a fact that it is common practice in high producing directorates that what the primary examiner decides is endorsed by the two other members, it confirms that simply having the production target in mind is not rendering a service to the public at large.

  10. Quality, what do you expectTuesday 12 November 2019 at 15:21:00 GMT

    I have to disagree as betting that no opposition will come is a mind set up which I cannot consider wise or as an example of good practice.

    If it is thus to make money by granting useless patents, I think this is a distortion of what the granting procedure should be.

    It is difficult to believe that all applications looked at by an examiner end up with a grant. If it is to close both eyes in order to achieve a given production target, I might show some understanding in view of the present situation at the EPO, but I cannot consider this a being correct.

  11. Thank you Quality, what do you expect. In the UK our Supreme Court has decided on a very generous doctrine of equivalents applying to the infringement scope. Essentially anything that relies on the same technical effect as a novel and inventive claim is covered by the claim for infringement purposes, with the exact factual situation (knowledge of the potential infringement) being used to determine what the equivalents are. So the scope of claim granted by the EPO is simply a guideline, i.e. the literal scope is the minimal scope (starting point) for analysis of what the claim covers. A huge amount of argument is going to go into what the infringement scope is during the proceedings. Doesn't it now make sense to let the EPO grant a minimal scope which is not ridiculous and let people fight it out in the courts on the few occasions a dispute arises?

    For the record, I also believe that your position of the Examiner refusing cases is valid and sensible. I just think it does not reflect commercial realities.

  12. Dear Sing-A-Pore,

    That the UK Supreme Court wanted to go away from the famous Improver decision is one thing. The pemetrexed case is however anything else than a model. On the one hand it is a premium for slapdash drafting and one the other hand, it has made a founded FTO study nearly impossible.

    It is true that in the new protocol on Art 62 after EPC 2000, the notion of equivalents was introduced, but during the diplomatic conference, the member states refused to endorse the definition of equivalents proposed by the EPO. So it is very easy to talk about equivalents, when everybody can understand w2hat he wants under the topic.

    That on top of it Lord Neuberger found it necessary to criticise the examiner for having raised an objection under Art 123(2)is neither fair nor correct. This attitude shows a clear lack of knowledge of EPO procedures and case law.

    The applicant only had examples for pemetrexed disodium in his application, but wanted originally to claim the association of any antifolate with vitamin B12, then pemetrexed in general, without the faintest support for it. The examiner had no choice but to raise a problem of Art 123(2), what was claimed was not supported by the original disclosure.

    By the way, the court of 1st instance in the Netherlands was not impressed by the behaviour of the applicant during examination and refused to consider that anything else than pemetrexed disodium was originally disclosed.

    At the EPO, the applicant wanted a quick grant, and never tried to bring in any proof that any salt of pemetrexed could work. He could have filed further experimental data or even file a divisional application to try to argue for a broader scope of protection. Nothing of this kind was brought forward or even envisaged.

    I doubt that the fathers of the EPO wanted it to merely grant patents being simply a kind of guideline and the rest to be fought out in national courts.

    Patents should be granted for improvements over the prior art, and not merely for some kind of possible guideline. If the EPO would do this, then it would price itself out of the market, as it cannot ask the price of a Rolls and deliver a kind of Mini having merely 4 wheels and some doors, and that the actual finishing touch is debated in front of national courts.

    The EPO should also publish a study to see how many patents are nullified or severely limited in front of national courts. The presumption of validity still applies, but it looks more and more like a giant with very fragile feet.

    Even if only 5% of patents are opposed before the EPO, in 2/3 of the cases the patent comes out battered from an opposition procedure (roughly 1/3 revoked, 1/3 maintained in amended form, i.e. limited) and only in 1/3 of the cases the opposition is rejected. Also something to think about when discussing quality. A linear extrapolation is certainly inappropriate, but there is food for thoughts.

  13. Thank you 'Quality, what do you expect' for your detailed reply.

  14. I wish to emphasise how unfortunate the pemetrexed decision by the UK Supreme Court was. The global titan pharma firm that applied for a patent on pemetrexed + B12 overlooked to include in the original patent application any announcement of that invention, or claim to that invention, at that level of generality. Inexcusable. And yet the Supreme Court excused that catastrophic failure to say, in the patent application as filed, what the invention was, for which protection was sought.

    Typical practice in the USA. But is that what we really want, everywhere else in the world?


All comments must be moderated by a member of the IPKat team before they appear on the blog. Comments will not be allowed if the contravene the IPKat policy that readers' comments should not be obscene or defamatory; they should not consist of ad hominem attacks on members of the blog team or other comment-posters and they should make a constructive contribution to the discussion of the post on which they purport to comment.

It is also the IPKat policy that comments should not be made completely anonymously, and users should use a consistent name or pseudonym (which should not itself be defamatory or obscene, or that of another real person), either in the "identity" field, or at the beginning of the comment. Current practice is to, however, allow a limited number of comments that contravene this policy, provided that the comment has a high degree of relevance and the comment chain does not become too difficult to follow.

Learn more here:

Powered by Blogger.