EPO: Lies, Damned Lies and Statistics

Guest Post by Peter Arrowsmith. Arrowsmith is a patent attorney and partner at the London based firm Cleveland IP

Amid great fanfare the European Patent Office has recently released its annual report, proclaiming another record year for the office: an all-time high in the number of filings and a 2.8% increase on the 2012 figures. These figures have been quoted widely in global media, and they create a positive impression about innovation and the success of the European patent system.

On closer inspection of the underlying data, however, a different picture emerges. It is apparent from these data that the EPO base their claims on the number of direct European patent applications added to the number of international-phase PCT applications, whether or not these international applications subsequently enter the European regional phase. This is immediately striking as a very strange measure for the number of European patent applications. Technically the figures from the EPO are not incorrect because Article 153(2) EPC says that an international application that designates the EPO shall be equivalent to a regular European application. However, the figures are misleading because the majority of international-phase PCT applications do not enter the European regional phase.

A more meaningful measurement of the number of European patent applications would comprise the number of direct European patent applications added to the number of PCT applications that enter the European regional phase. By this measurement it is apparent that the number of European patent applications actually decreased by around 0.5% in the last year. Also, the total number of applications is more than 100,000 less than the number quoted by the EPO. In fact, it would appear that the number of European patent applications filed reached a peak in 2010 and that there have been significantly fewer applications in each of the following years.

The President of the EPO blogged about the statistics on 12 March 2014, saying: “The filing figures for 2013 confirm the trends of the past few years, with an overall increase (266 000 filings, +2.8% versus 2012) resulting once more in an all-time high”. Although not technically incorrect, this statement from the President is, at best, misleading. At worst the EPO are engaging in a wilful misrepresentation of the facts.

Of course, the EPO has an interest in promoting its own role in stimulating innovation. Therefore, it is not surprising that the organisation attempts to put a positive spin on facts. However, it is troubling when this spin is relied upon by governments, industry and individuals when it comes to important decisions on the patent system.

35 thoughts on “EPO: Lies, Damned Lies and Statistics

  1. 10

    However, it is troubling when this spin is relied upon by governments, industry and individuals when it comes to important decisions on the patent system.

    Hmmm, maybe those who know this and are bound to some type of ethics to the law should NOT be spinning (even on places like blogs).

    Yes Virginia, intellectual honesty is important on blogs too.

  2. 9

    Regarding EPO statistics, the comparison with past annual reports is telling.

    The 2011 EPO annual report mentioned separately the number of EPO direct applications and the number of PCT applications entering the EPO regional phase.

    Since 2012, the EPO annual report only discloses the addition of these figures.

  3. 8

    Quite right The Critic, as Mr. Twain was bemoaning the lack of protection at the time (imagine, wanting to make money off of his thoughts captured on paper. Maybe he was just a bottom feeding grifter…)

    (link not being accepted)

    1. 8.1

      I can’t remember where, but I am sure that Mark Twain wrote extensively about how unfair it was that his writings were being copied and he was not being compensated for it. It has been 20 years since I read his writing, but I know it is there.

  4. 7

    Question 1: Was there a huge recession sometime in the last 5-10 years or so that might have impacted EPO patent filings?

    Question 2: Does anyone in the EP have any data suggesting that annual “all-time high” rates of patent filings correlate with anything other than “all-time high” rates of patent litigation?

    1. 7.2

      …because everyone knows that the only thing that is the worst thing EVER of there being patents is enforcing patents…

    2. 7.3

      Q1. – Well I believe the banks did screw up the economy in 2008. Although filings in the rest of the world have recovered, they have not not in Europe.

      Q2. – Traditionally the EPO have not been concerned with litigation and so ignore the effect of what they do. They are barely capable of counting what they have, and so don’t expect too much in the way of analysis.

      1. 7.3.1

        Q1: not sure about Europe but in the U.S., to the extent banks “screwed things up” it was in predictable response to incentives passed by Congress (called “CRTA” laws over here) to loosen lending standards for putatively underserved sectors of the population. Knowledgeable Americans thank Barney Frank and Chris Dodd, not Citi.

  5. 6

    “At worst the EPO are engaging in a wilful misrepresentation of the facts.”

    Since when would anything like that surprise anyone concerned with a patent system?

    1. 4.1

      Quite so, Meldrew. Focus.

      The way it has been explained to me, in China the investmentworthiness of a start up is judged by its portfolio of patent filings, not on the validity and reach of the claims that have gone to issue. So they all get as many WO publications as they can, get the state to cough up the money, then abort any further loss of their precious capital down the sink of further prosecution of their A-published junk patent applications.

  6. 3

    Mindful of the prior art provisions of the AIA I think we might reflect on which filing figures are more meaningful for US readers; the number of EPO national phase entries or the number of WO publications, for the EPO stats are actually telling us that the number of WO publications is indeed increasing.

    I have in mind why people go to WO publication, especially those in the PRC.

    For whatever reason, applicants in China file an awe-inspiring number of PCT applications (and so get given an EPO filing serial number) and prosecute them all through to WO publication. But then, for whatever reason, they carry a vanishingly small proportion of those WO publications on into the EPO national phase. Ergo the Arrowsmith statistical point.

    Applicants in Asia are fond of using Patent Office A publications as a way to keep their technological forward path open. It’s a legitimate use of the patent system, isn’t it? So increasing WO publication numbers do somehow show that innovation is still accelerating. Bravo!

    1. 3.1

      and prosecute them all through to WO publication.

      Um, how much prosecution effort is involved in that?

      1. 3.1.1

        Indeed, as far as I was aware, only passively receiving a PCT WO publication and a preliminary prior art search. Typically then followed by the client making a sensible decision to save a great deal of money by NOT filing in the EPO and other foreign countries (after seeing those search results) during the long time delay which the PCT filing provides?

    2. 3.2

      So increasing WO publication numbers do somehow show that innovation is still accelerating. Bravo!

      MaxDrei – serious question – Does the EP have an analogous concept to the US Quid Pro Quo, and if so, how is the implication inherent in your statement that one side of that balance is being enriched with more and more publications resulting in a an equal (and balanced) enrichment of the applicants?

      1. 3.2.1

        Serious answer anon: I do not understand your question so cannot begin to answer it.

        But every patent system has at root the same quid pro quo: if you enable in your patent application a new, useful and not obvious contribution in a field of technology, you are in line for the grant of 20 years of exclusive rights.

        Or you can opt to keep it a trade secret, for nobody forces you to file.

        The bargain is given regardless whether of not your patent application is laid open only after issue or earlier, 18 months after filing.

        The quid pro quo is fairer in the rest of the world than in the USA under the AIA because in the rest of the world obviousness attacks can only be run on the basis of what was made available to the public before the filing date of the patent application. All that prior art you can search and assess long before the cut-off date for pulling the pending patent application to prevent its A publication. So you can do a comprehensive search which will tell you whether or not you can get to issue, and then decide at leisure, long before you get to the cut off date, to pull your patent application to prevent its A publication.


          I do not understand your question so cannot begin to answer it.

          That’s pretty funny, as you then proceed to begin to answer it.

          First part: “But every patent system has at root the same quid pro quo

          OK – that answers the first part of the question in that not only does the EP have a Quid Pro Quo, you feel that the Quid Pro Quo is the same and serves as the root of all patent systems.

          Tell me then, in order to makes things simple for you,

          What is the Quo?
          What is the Quid?

          Take this in easy bite-sized answers. First, the Quo. What is it that the patent office (and as proxy for society) get?

          After you answer this first part, then we can get to the second part.

    3. 3.3

      Max, re China. I think the reason the Chinese use the EPO for searching but without going into actual prosecution is because the EPO is probably the best in the world for searching, but the cost of prosecuting and maintaining European patents throughout Europe is staggering.

      Thus, if they go forward, they might go forward in China, perhaps in other Far East nations and in the United States.

      1. 3.3.1


        The vast majority of Chinese origin PCT applications are searched by China as ISA.

        As for where they enter the national phase, USA is the main destination followed by EPO, China, Japan, India, Korea – in that order.

        Interestingly, the national phase rate appears to be declining (for Europe at least).

        Note, since AIA made published PCT applications prior art as of the day of filing, wherever filed and in whatever language, expect PCT filing rates to go up, simply as US defensive publications. Max Drei’s “clearing the path” explanation for Chinese filings has a lot of merit.


          MaxDrei had once started a discussion on the treatment that has been exasperated with the AIA expansion to PCT applications (as you mention) for 103 obviousness effect, but departed from that discussion when the discussion turned to describing why the notion of obviousness pertains (correctly as it does) to the legal construct of PHOSITA, instead of to a live flesh and blood person.

          This notion is still inherent in MaxDrei’s post 3.2.1 and the notion that the attack on the patentee is somehow misconstrued to make the patentee somehow be in the place of the PHOSITA. Of course, this may be due to the fact that the US PHOSITA is a super-charged omniscient “ordinary” skilled person knowing the state of the art of any and all items legally deemed prior art (and thus including filed but not yet published items).

          Tis not a big leap for the super-charged PHOSITA, given that this “ordinary” person is credited with knowing ALL published items prior to the critical date anyway (something hardly ordinary, eh?).

    4. 3.4

      The panglossian approach to EPO statistics is not new. See ipnoncredere where a number of posts going back to January 2013 comment on the pointless numbers the EPO trumpets.

      The Chinese issue you raise is probably related to poorly directed subsidies for overseas filings. See for example patentology.

      What can be seen from WIPO statistics is that in 2012 there were 60% more US national phase filings from Chinese PCT applications than there were EP regional phase entries from Chinese PCT applications. The US is more attractive than EP for Chinese applicants. Is that because the Chinese are fond of junk patents?

  7. 2

    The EPO is not alone. Consider that the internal subjective gauges of “quality” at the USPTO have for years routinely shown 95% and above “compliance” in both final rejections and allowances, but that a far more objective measure – what three ALJs think of final rejections in cases cherry-picked by examiners to proceed to appeal – shows that almost half (usually between 42%-48% year after year) of final rejections suffer “clear legal error” (the standard at the Board) in one or more respects (reversals + partial reversals). And note that in appealed cases the examiner enters with a presumption of correctness on his side. It is up to the appellant to demonstrate clear error.

    While the PTO’s internal grades of its final dispositions remain enormously self-congratulatory, its “composite” quality score hovers around 70% because 25% of the composite total accounts for two unsupervised inputs – evaluations from practitioners, and evaluations from examiners, neither of which, let’s just say, tend to flatter the PTO. Remarkably, the “composite” score steadfastly refuses to acknowledge the only objective indication extant for final rejection quality – the almost 50% loss ratio at the Board, steady as a rock year in year out.

    Several years ago I pointed this out an assistant commissioner. He complacently told me that Board reversals were backward-looking, and that the (same) 50% rate back then would decrease over time as the then-new “quality” programs that were resulting in the subjective 95% numbers kicked in. That never happened. But the “composite quality” thermometer gauge in the PTO annual reports inexorably rises.

    1. 2.1

      Let’s not forget the Reject-Reject-Reject era of what “quality” was defined as and that it took Kappos to directly refute that with his historic statement of “Quality does not equal reject.”

    2. 2.2

      Plus counting U.S. RCE’s and other mere paid continuations of prosecution of the same applications in statistics for both new applications and abandonments?

    3. 2.3

      Tour, the 2010-2015 strategic plan requires the patent office to “measure and improve patent quality.” If one looks at their dashboard on quality, there is no measurement of patent quality other than a survey. What questions are on that survey is not known. But as we all know, there is a widespread complaint that the patent office is issuing an enormous number of low-quality patents that are really hurting the American economy. There was one statistic cited in one of the briefs that fully half of all patents issued today are so-called software patents. These patents are notoriously of low-quality, if not every one of them being invalid for claiming ineligible subject matter. Yet the patent office seems totally unconcerned, measuring patent quality by internal performance standards, not by whether the product is anything other than trash.

      The movie “The Bridge on the River Kwai” portrayed an event that happened in a Japanese prisoner of war camp somewhere in the jungles of Southeast Asia. A bridge needed to be built to carry the Japanese war effort into India. A Japanese commander couldn’t get the bridge built because his troops did not have the requisite technical capability, and because the British troops they commanded were resisting. He then hired the British commander, putting him in full charge of not only the British troops but the Japanese troops. The British commander took it upon himself to show the Japanese just how good the British Army was. He made it is objective to build a high-quality bridge on time. His subordinates questioned whether this was wise. He dismissed their questions telling them that doing a good job was important for British morale.

      He completed the bridge on-time, and it was high-quality. However, the British Army, having learned that the bridge is being built, sent in a team to blow it up. Had they not done so, the implication was that the Japanese could have won the war.

      So what was a lesson here? The lesson seems to be, from the British point of view, that creating a highly efficient workforce, with high morale, while not keeping the eye on the ultimate objective is a plan for ultimate failure.

      It seems quite obvious to anyone that the patent office is not doing a good job, and has not been doing a good job for a very long time, in terms of patent quality.

      1. 2.3.1

        It is true that the movie showed the bridge to be destroyed; however, the original novel by Pierre Boulle did not have that outcome: the saboteurs were eliminated before blowing up the bridge. That was the irony of the whole matter, and it was completely lost on the altar of feel-good.

  8. 1

    Dennis, for the benefit of any oversensitive EPO readers, your blog title here of “Lies, damned lies, and statistics” is a humorous phrase made popular by Mark Twain and others, including the title of a 1978 book by Michael Wheeler.

    1. 1.1

      “Only one thing is impossible for God: To find any sense in any copyright law on the planet.”

      -Mark Twain

      1. 1.1.1

        “Only one thing is impossible for God: To find any sense in any copyright law on the planet.”

        I suspect that this quote is being taking out of context by the anti-copyright crowd.

        Mark Twain was very much for copyright protection. One of his books was being circulated in US — before the US release — because of counterfeit publishing within Canada (perfectly legal by Canadian law). Henceforth, Twain reportedly took an active interest in Copyright laws.

        At the time, it was common for books published in a first country to be free republished within other countries. Also, the term of copyrights was considerably less than today.

Comments are closed.