National filings: bad habit or necessary virtue?

Earlier this year, as part of its Balance of Competences Review, the UK's Intellectual Property Office published a document, Review of the Internal Market: Free Movement of Goods; including the EU Customs Union and Intellectual Property Rights: Call for Evidence.  On page 16 of this document there appears a table, reproduced below, that caught the eye of one of the IPKat's readers:

The table would appear to indicate, among other things, that the UK's national patent system was pretty popular with UK applicants; this was nothing, however, to the affection for the jurisdiction's trade mark system among local applicants. And while UK registered design filings are not a vast number, local applicants are practically the only applicants.

What does this table reflect? Does it show a hopelessly introverted, domestically fixated approach to IP protection on the part of UK businesses? Or is it that the national application is the first resort of that fabled creature which all governments purport to worship and which doesn't have a voice of its own -- the small and/or medium enterprise, or SME?

The IPKat wonders whether comparable figures in other European Union Member States show similar trends: if it does, then perhaps the Brits aren't so uniquely insular after all.  Says Merpel: Community trade mark and Community registered designs haven't shaken Brits free from the habit of local filing. Is that, then, why the European Commission has been so anxious to install a Community-wide patent system to replace national patents rather than to compete with them, so that another little sphere of national influence in supra-national Europe can be removed?
National filings: bad habit or necessary virtue? National filings: bad habit or necessary virtue? Reviewed by Jeremy on Monday, August 19, 2013 Rating: 5


  1. At least one factor may be that the UKIPO search is very cheap, and considered (at least by UK practitioners) to be very thorough, making it a good choice for priority filings. This lets the applicant look over the prior art before filing the EP/PCT application.

    This doesn't really explain the high percentage of local applicants though.

    I'd also question whether "applications received" includes PCT applications filed via UKIPO, which would significantly inflate the numbers

  2. The Germans make substantial use of their national system. Indeed, many German companies still file both German national and European patent applications. Germany still has the seven-year examination request period, and by the time seven years have passed, you know whether you want it or not. It allows German applicants to consider which set of claims is better and therefore to validate or not validate a granted EP patent in Germany.

  3. Would I earn the cat's meow if I were to remind everyone that patent rights are strictly territorial?

    As such, the phrase "hopelessly introverted, domestically fixated approach" seems hopelessly misplaced.

  4. The cheap and rapid search is the obvious reason for first filing in the UK.

    What would be informative is the proportion of cases for which no response is ever made to the search / combined search / exam...

    And as for local applicants being almost the only applicants - I would say 'well, d'uh' - overseas applicants will enter via Europe to secure at least DE, FR and GB, representing some 70% of EU GDP between them. Only in very niche areas (so niche I can't offhand think of any) would someone from Japan, the US or Korea want a UK patent but not a DE or FR one...

  5. German attorney firms make substantial use of the UK national system, as well. A GB first filing has a good reputation with many DE attorneys, at least to for the benefit of the very cheap but reasonably good quality search, performed quite promptly.
    Under the assumption that a quick result is desired.

  6. First comment. Your statistics may be distorted by variations in national laws.

    A few EPC member states oblige their nationals and/or residents to file domestically their priority applications.

    The most notable example of these is France. Foreign first filings by French residents are punishable by fine, or even by emprisonment if the disclosure concerned a defence-sensitive technology.

    I don't know whether these dispositions are actually enforced, but you'd be foolish not to make your first filing as a national FR application. The reason is that the French searches are performed by the EPO, but for a much lower fee. You get essentially the same search report and opinion which you would have got from filing directly an EP or WO application. If your second EP or WO filing is essentially identical with the FR first filing you also have your search fee reimbursed. All in all a pretty good deal, you'd be quite foolish not to use the system as it is.

    First filings of any kind have top priority at the EPO, and you'll have a good idea about patentability long before the Paris year is out.

    You can also get an EPO search (and also opinion in some cases) for some other countries, including BE, IT, TR, GR, and NL.

    The EPO only began a few years ago to search Italian applications, I believe that it might have something to do with the (re)activation of IIB agreements. Applications are forwarded to the EPO in the original Italian, together with a (wretched) machine translation of the description and a human translation of the claims. If you can grok Italian you're better leaving these aside, and concentrate on the original documents. The opinion is drafted in English.

    The occasional BE and NL applications are processed directly in Dutch the unofficial fourth EPO language.


  7. RTF states: "if the disclosure concerned a defence-sensitive technology"

    Similar restrictions apply to the U.S.

  8. Jeremy,

    In your table, if the number of trademark and design applications usually represent effective attempts to secure national IP rights, the same cannot be said for patent applications. Most of them never mature into anything, or merely serve as priority applications for foreign second filings.

    The number of 22,000 is therefore misleading, albeit formally correct. An application may be validly filed without paying a search fee or a publication fee. A more useful and accurate number is the number of A publications in a given year, as applications which are maintained and pass that milestone are much closer to being real ones. Roughly about 70% of "A" GB publications mature into "B" patents.

    From Espacenet, and neglecting time-lag effects, the number of "A" publications for the year 2011 is about 10,000 - only 45% or so of the number quoted.

    British patents are thus nowadays even more marginal than suggested by the table.

    Here is an historical plot of the number of British patent publications over the last 150 years. It is mostly derived from a compilation of old patent numbers published by UKIPO.

    The solid lines before 1979 represent mostly granted patents (with the exception 1907-1949 during which refused foreign patents were published anyway).

    Both world wars caused a sharp drop in the number of patents granted. The gap at 1916 corresponds to a modification in the numbering system.

    After 1978 all applications were published before grant, so the red line cannot really be compared to the other data. The magenta data points were obtained by me from Espacenet to complete the series after 2010, and match well the data from UKIPO. (The web site now has data points for the latest years too. I had commenced this plot sometimes in 2011.)

    The cyan line represents a rough and ready approximation of the number of patents which should have issued from the published applications in the red line (70%).

    Two great changes occured in the British patent system over that period:

    1) The Paris Union. The jump after 1883 is quite remarkable. (Incidentally, the overall grant rate dropped from 2:3 to 1:2.)

    2) The accession to the EPC. The collapse in British applications after 1978 is remarkable, but not quite as brutal as in other countries which became members later. The users were probably slow in adopting a still untested system.

    In absolute terms, and setting aside the cataclysms of both world wars, the number of titles granted by UKIPO dropped around 1990 to a level never seen since 1883!

    In relative terms, considering that the population more than doubled since the 1880s (give or take Ireland), and that the GDP increased more than sixfold since the same period (whatever such numbers mean), one must see that the UKIPO has pretty much lost most relevance.

  9. What is a UK Applicant? OMG 31% are from outside the UK!

    The UKIPO relevance depends on the interest in IP of UK Applicants. It is IP that has pretty much lost its relevance to UK business. That is what the stats, poor as they are, show.

  10. Just a couple of points:

    * While foreign applications that did not proceed to acceptance were indeed published (in the state the file was at 18 months from priority), after about 1941 they were no longer printed or given a serial number in the published patents series, due to wartime paper shortages. Publication of such applications was by file inspection only, and their existence would only have been discoverable from the information on new applications published in the OJ(P) some 6 weeks after filing.

    In fact all applications of that era were laid open to inspection in their "18 months" state: some abstracts from that period include descriptions and depictions of embodiments that were deleted during prosecution but which were present in the file as "open to inspection", the abstracts carrying a note to the effect that the relevant matter does not appear in the application as accepted. The prosecution files of such applications have long since been destroyed, and only the abstract information remains.

    * Germany has long had a rapid grant procedure available for non-convention applications, but few people outside Germany seem to be aware of the fact. When I used to write patent abstracts for "Derwent" in the late 1970's and 1980's, about 4% were "Patentschrifts" published less than 18 months after filing. I think the earliest I saw was published 7 months after filing. No special fee apparently, it was/is only necessary to pay all normal fees up front on filing, and if all was/is in order, you had/have the option of prompt grant.

  11. Ron,

    Thanks for the feedback, which fuelled the nagging doubt I had about the period 1907-1949. I obtained some more data which I massaged until they purred, and here are the results. Spoiler alert: it's not a problem. (My hunch was that it wasn't either).

    I requested from the EPO OPS server 700 cover pages from the period 1916-1940 in three batches of 350 each. (I did not try to obtain documents for the period 1907-1915 because of the previous document number format entailed too much work.)

    The document numbers of first batch one were selected at even intervals in the range from GB102223 to GB529000 inclusively. The second and third batches used the numbers of the first batch augmented by the magical constants 57 and 757.

    Each run succesfully returned 346, 346 and 347 pages respectively. The coverage is quite good. I cannot offer any hypothesis about the the missing points. It probably doesn't matter much.

    I then flipped through the cover pages, scanning visually for the expression "complete not accepted".

    Each batch contained 8, 5, and 10 documents respectively, for a total of 23 out of 1039.

    All 23 documents in questions claimed a convention date, and carried a note at the top on the lines of:

    Note .— The application for a Patent has become void. This print shows the Specification as it became open to public inspection under Section 91(3)(a) of the Acts.

    The proportion of documents published over the period 1916-1940 disclosing not accepted foreign applications rather that accepted ones is therefore about 2.2%, ±0.9% @ 95% confidence.

    About 4 further documents out of the sample carry a stamp to the effect that a accepted application (of domestic or foreign origin) was published despite being void, apparently for the reason that the final sealing fee hadn't been settled. These aren't counted in the above.

    Of the set of accepted applications, 28.5% (±2.7% @95% confidence) claimed a foreign priority. The same cannot be said today: the majority of rights held in any EP country are most probably detained by an owner residing in another country.

    The contribution of void patents publications to the period 1907-1949 has an almost negligible effect on the data. The number of publications is so small that it raises a suspicion. On what criteria did the patent office decide to publish a foreign application?

    The large difference between the number of reported filings (22000) and the number of publications (ca. 10000) puzzled me. This table from the latest UKIPO annual report seems to shed a little light. Filing fees are paid for about 80% of the applications, and most of them are searched. The greatest attrition appears to come after the search report is issued, where the applications are abandoned. My experience with EP-direct applications is that a withdrawal before publication is relatively rare, even after a crushing search report. This would suggest a difference in behaviour or strategy between UKIPO and EPO users.

    This doesn't change IMO my initial conclusion, but the picture is a bit more complicated than I initially conceived it.

  12. The Patent Office didn't need to exercise any discretion. The obligation to publish convention applications in their "open to inspection" state at 18 months ( which was the normal period for putting an application in order at that time) was enshrined in the respective Patents Acts of that period.


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.