Has the Global Patent System Weakened in the Last Decade? Assessing the Strength of National Patent Systems

By Gordon Harris, Alexandra Brodie, and Jamie Rowlands

Published in Landslide Vol. 11 No. 4, ©2019 by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association.

The patent system in the United States has been impacted by changes introduced in the last dozen or so years, which are seen by some as having weak ened the standing of the U.S. system compared with patent systems in other major economies.

Getty Images

The difficulty with comparison is: what do you compare? Absolute patent application or grant numbers? Growth in patent filings? Strength in litigated patents? Commercially meaningful outcomes? What is the meaningful data? These can differ from sector to sector, and certainly from jurisdiction to jurisdiction. Always it is important to understand what the data drawn upon actually reflects. Nevertheless, in this article we have a go.


When considering whether innovators see strength in the patent system, perhaps the best data to review concerns the number of application filings.

Worldwide, patent application numbers have been rising steadily since the turn of the millennium; between 2002 and 2016, world application numbers more than doubled, to over 3 million.1

European patent applications have increased fairly steadily in the decade to 2018, from 145,557 in 2008 to 165,235 in 2017 (a 12.0 percent increase).2 Of those filings, 27.8 percent are classified as “electrical engineering”; 26.0 percent as “chemistry” (including pharmaceuticals and biotechnology (7.8 percent)); and 22.0 percent as “mechanical engineering” (including transport (5.0 percent)). Within those fairly steady headline percentages, in electrical engineering there was steady real and percentage growth in digital communication filings between 2008 (4.4 percent) and 2017 (7.1 percent), along with reduction in telecommunications filings (4.1 percent in 2008 to 2.3 percent in 2017). Relatively static numbers of filings in the pharmaceutical and biotechnology classifications reduced the percentage contributed by those sectors from 8.2 percent to 7.2 percent of all filings.

Chinese patent applications have increased exponentially in recent years, from 153,060 in 2007 to 1,381,594 in 2017 for invention patents, and from 179,999 in 2007 to 1,687,593 in 2017 for utility model patents. This amounts to approximately a ninefold increase for both. Of those filings, in 2017, 30.5 percent are classified as electrical and electronics (including telecommunications and computer engineering); 28.9 percent are classified as manufacturing and transportation; 19 percent are classified as mechanical engineering (machines, structures, etc.); and 8.8 percent are classified as chemistry and metallurgy (including pharmaceutical and biotechnology). There was a dramatic increase in the manufacturing and transportation sector from 16.9 percent in 2007 to 28.9 percent in 2017.3

However, patent application data represents only part of the story. An application only results in a patent if it proceeds to grant. In Europe, the landscape is particularly complicated in this respect. First, the applicant for a European patent must choose the countries4 (“designations”) in respect of which a decision to grant the patent will apply. In many countries there is then an additional “validation” stage necessary at the national level, involving filing of translations, in order for the relevant European patent to be enforceable in that country. Available data indicates significantly higher validation rates for Germany (over 90 percent) and France and the United Kingdom (both 70–90 percent) in the first decade of this millennium than for other European Patent Convention (EPC) states.5 A European patent may also be allowed to lapse at different times in different countries. Second, applications for patents of national scope may be filed, in addition to or instead of European patent applications; and decisions about whether to grant either or both (for some jurisdictions) may be postponed a few years to the time of grant, which considerably complicates analysis of application data. This is the case particularly at the present time, in view of the uncertainties regarding the proposed introduction of the Unified Patent Court (UPC) and unitary patent system.

Accordingly, application data (and granted patent data) can at best indicate only the perceived enforceability of patent monopolies granted previously and of the anticipated commercial value of holding patent rights in future years. These are impacted by many factors, the balance of which can differ across different areas of technology and different geographical regions.

Post-Grant Validity Proceedings

While patents can usually be invalidated by the courts of the relevant country following litigation at any point in their term (generally 20 years from filing), many patent systems provide a post-grant procedure for challenging the grant of a patent without recourse to the courts. Such office and court outcome data can differ.

For example, in the United States, data6 suggests that more than half of tech patents submitted to inter partes review (IPR) are found invalid, a finding of full validity having been reached in less than one-fifth of IPRs since 2014. Pharmaceutical patents have fared better, with 46 percent found invalid and 28 percent found valid.

European patents may be challenged in opposition proceedings filed within nine months of the patent’s grant. As seen for the United States, the data indicates a trend toward invalidity in recent years for tech patents, but a less marked change than in the United States: between 2000 and 2014, full validity was found in 29 percent of tech cases, whereas for 2015 onward, the rate is 24 percent. In pharmaceutical cases, the respective data indicates validity in 37 percent of cases for the years 2000–2014, and in 40 percent of cases since 2015. However this shift has been broadly matched with a reduction in partial validity findings, such that the invalidity rate remains broadly consistent.

In China, patents can only be invalidated by the Patent Re-Examination Board, which is part of the Patent Office. Validity cannot be challenged in the civil courts, whether or not related infringement proceedings are on foot. The Patent Re-Examination Board received 4,565 requests in 2017 for an invalidation declaration, an increase of 15 percent from the previous year. However, reliable data on outcome is not yet available.

Court Proceedings

Analyzing outcomes in court proceedings is more complex. When it comes to enforcement, validity (or partial validity) alone is not enough; there must also be infringement of a valid claim. Different legal systems handle the interrelationship between these issues in different ways. For example in China, Germany, and some other jurisdictions, “bifurcated” issues of infringement and validity may be determined by different courts in separate proceedings. In common law systems, both issues tend to be heard in a single set of proceedings; in any particular case, one issue may not reach proceedings or may fall away before trial. This makes comparison of data collected from different jurisdictions quite difficult.

In U.S. court proceedings, data7 since 2000 indicates greater success for patent owners than in IPR proceedings, with data indicating that in the years 2000–2014, full validity was found in 57 percent and partial validity in 22 percent of cases. The data for 2015 onward indicates that the trend in the courts has moved against the patent owner, to full validity in 35 percent and partial validity in 9 percent of cases for the years 2015–2018. While the absolute data drawn upon is difficult to scrutinize, the trend observed is interesting. In pharmaceutical patent cases, the data indicates full validity being found in over 60 percent of cases over the years 2000–2018, and broadly consistent rates for findings of full validity for the earlier and later periods (at 64 percent and 65 percent for 2000–2014 and 2015–2018, respectively).

In the United Kingdom, data8 for 2015 onward indicates that in cases which reached substantive judgment in England and Wales, the patent owner was successful 40 percent of the time in establishing infringement of a valid claim or in defending a challenge to validity (at least in part) where infringement was not an issue determined by the court. In a further 5 percent of cases, the patent was at least partly valid but not infringed on the facts. For tech patents, the success rate for patent owners is 44 percent, and for alleged infringers/challengers, 50 percent. For pharmaceutical and biotechnological patents, the success rates are 34 percent for patent owners and 61 percent for alleged infringers/challengers.

Interim Relief

In some technological sectors (particularly the pharmaceutical sector) and in some jurisdictions, the availability of preliminary relief to restrain patent infringement pending the court’s substantive judgment forms an important component of the commercial value of a patent.

Germany is a jurisdiction in which obtaining preliminary injunctive relief is relatively straightforward in cases involving noncomplex technology, clear-cut infringement, and a patent which has survived opposition or nullity proceedings. The award may be made on an ex parte basis or following a hearing scheduled by the court, should the court consider it necessary to hear the submissions of the alleged infringing party.

In contrast, in the United Kingdom, except in unusually urgent scenarios (e.g., where the patentee contacts the court within hours of becoming aware of a very imminent threat of infringement9), where the patentee seeks interim relief and the question of its award proceeds to consideration by the court, the balance of the parties’ interests usually proves the deciding factor. Case data for the United Kingdom therefore records a lower rate of award of interim injunctive relief than in Germany. This may in part be offset by practice in the United Kingdom for alleged infringers to agree to an interim injunction before the issue reaches determination by the court. Nevertheless, the ready availability of without notice interim relief in Germany may contribute to the higher validation rate for European patents in Germany compared to the United Kingdom (noted above).

Preliminary injunctive relief in China for invention patents and/or utility models is very rare. Despite the fact that the Supreme People’s Court produced a judicial interpretation in 2015 which, among other things, set out the criteria that courts should consider when hearing preliminary injunction applications, successful applications remain very low. There are a number of reasons for this, but Chinese courts are reluctant to support the requirement of “irreparable harm,” and there remains a seemingly very high threshold for the claimant to show that the case will succeed at the substantial trial, thereby creating a landscape whereby unless the case is a “slam dunk” it is unlikely to prevail at the interim phase.

Final Relief

The availability and form of final relief also impacts the enforceability and commercial value of patents.

In the United States, a statute permits the courts to increase the damages up to three times the amount found or assessed,10 a relevant factor being what the infringer knew at the time of the infringing activity.11 Historically, in the United States, upon a finding of infringement the courts have granted injunctive relief in the vast majority of patent cases.12 Yet as an equitable remedy, the award will only be justified where the four factors of the eBay test—irreparable injury, inadequacy of monetary damages, the balance of hardships, and the public interest—are met. Nichia v. Everlight Americas is an example of a case in which the test was not met, a factor in the balance being Nichia’s willingness to license its technology to other competitors.13

In the European Union (EU), punitive damages are not permitted. Harmonizing legislation on relief for infringement14 mandates forms of relief that national courts must have power to award following a finding of infringement, as well as the principles governing the exercise of the courts’ discretion. Awards of relief must be fair and equitable, effective, proportionate, and dissuasive. Damages must be “appropriate to the actual prejudice suffered by him/her as a result of the infringement.”15 Broadly speaking, this is considered compensatory, but there is some scope for moderate uplift in cases of flagrant infringement. Perhaps more notable is the variation that remains between different EU member states regarding the proportionality aspect of the test for injunctive relief, and accordingly the availability of injunctive relief, as well as its shape and conditions, particularly in cases involving challenging ethical contexts.

In Germany, for example, the national patents statute does not confer upon the courts discretion in respect of the grant of injunctive relief. While the requirement for proportionality in the award of injunctive relief has been acknowledged by the courts in some cases, this is not seen to prevent the award of injunctions in practice.16

In contrast, in the United Kingdom, the common law system imbues the courts with broad inherent powers, which they employ to shape standard injunctive relief to achieve justice in the circumstances of any particular case. In the context of patents essential to a technological standard, the English courts have developed the “FRAND injunction” to restrain infringement in the United Kingdom unless the infringer accepts and abides by a license on fair, reasonable, and nondiscriminatory (FRAND) terms, as such terms are determined by the court, which may comprise a global geographical scope.17 The Patents Court has indicated that it would award an antisuit injunction to restrain a litigant from seeking to interfere with the freedom of the English court in this respect.18 The courts have also demonstrated sensitivity in the scope of injunctive relief awarded in life sciences cases: in order to enable measured transition to noninfringing alternatives, and to ensure that patient need is met where no noninfringing alternative is available;19 and to facilitate continuation of research and development projects.20 How to award appropriate injunctive relief for infringement of second medical use patents where some medical indications are outside the scope of patent protection remains a developing area of the law.21

In relation to final injunctions, Chinese courts will, almost without fail, use their discretion to award the successful claimant with permanent injunctive relief. This trend makes China an attractive jurisdiction to litigate even though damages remain low (although this trend is slowing changing).

There has been a steady rise in standard-essential patent (SEP) and FRAND issues being heard in China. Critically, courts in China are willing to grant an injunction against an implementer of SEPs22 and to take a pragmatic review of FRAND negotiations between the parties and ask the question on the facts “who is at fault for the breakdown of the negotiation?” If the implementer is at fault, an injunction can be awarded in China as well as damages (including punitive damages). If the SEP holder is at fault, no injunction will be granted. China is a civil code jurisdiction and, therefore, does not follow the doctrine of court precedent. This makes it difficult to see clear trends in the early stages of the SEP/FRAND issues. However, it is useful (although persuasive only rather than binding) that the Guangdong High Court released SEP Guidelines, which became effective in April 2018. These set out a broadly similar approach to the U.K. court in the Unwired Planet case (including, in certain circumstances, a determination of the license terms of patents outside of China).

This level of data analysis still remains only a part of the patent enforcement landscape, and thus only a partial indicator of commercial value.

Changes in the Pipeline?

Will the anticipated UPC and unitary patent system harmonize the patents landscape in Europe?

It is difficult to say. The new system is presently on hold, pending resolution of a challenge filed in the German constitutional court. Despite the United Kingdom’s ratification of the underlying treaties, it is also presently unclear whether the United Kingdom will be able to remain in the new system following Brexit.

If and when the new system does become operational, it will introduce a new court with jurisdiction to hear disputes regarding the infringement and validity of European patents in respect of the countries that have signed up to the new system. This is a subset of the European patent countries and a subset of the EU countries. Despite this, and although it may not include the United Kingdom and will not include Poland, the UPC would include the major EU economies of Austria, France, Germany, Italy, and the Netherlands. For the UPC countries, the new court would not replace the existing systems, at least for the first seven years of its operation. Instead, it would create another court with concurrent jurisdiction to the national courts in respect of European patents in their present form, and with sole jurisdiction in respect of unitary patents—a new form of European patent designation for the participating countries. Accordingly, the new court would enable a single forum for issues of infringement (including relief) and validity for the participating countries. This may streamline the hearing of patent disputes in the UPC countries. It may also facilitate complex and time-consuming jurisdictional battles which make the present system seem straightforward.


Meaningful comparison of the strength of one developed patent system against another is difficult. Application numbers represent a very crude measurement of the “success” of a system. Grant rates also. Are high injunction rates a measure of strength? This is not clear either: a strong patent system must enable invalid patents to be identified and enforcement measures that are clearly detrimental to the public interest to be refused. So at what level should we expect to see relief refused and patents invalidated in a “strong” patent system? The answer to this question can be difficult to detangle from the philosophies underpinning the differing legal systems within which the major patent systems sit.

What is clear is that for as long as the patent systems of the world are governed by national legal systems, as they presently are, an invention will only be protectable in a jurisdiction if a patent has been granted for that jurisdiction. A patent system may be seen as more or less inventor-friendly, by multiple different measures, but ultimately it is not substitutable by another system. The United States is a driving force in innovation and economic growth, so its patent system can expect to remain in demand.

Stepping back, it should perhaps be noted that the basic outline of the patent system is the same around the world—a highly examined monopoly lasting 20 years. That system derives from a time of much slower moving inventive pace, and has added complications and costs (office fees and translation costs), which make the obtaining of global coverage a time-consuming and very expensive exercise.

Is that level of expense, and the time taken to get to grant in some countries, compatible with the current pace of technological advance? Can companies at the cutting edge of artificial intelligence (AI) and connectivity be expected fund the process of keeping their patenting activity running alongside their technological innovation? To some extent the same can be said of the medical sector, where the ability to obtain suitable patent protection for molecular and biological pharmaceuticals was in line with the pace of change, but which has been joined by fast-moving biotech and genetic innovation and the overlap of technology and diagnostics.

Is a 20-year term necessary or useful now, other than to deal with the regulatory delays in getting some products to market? Could that issue not be addressed in other ways? Renewal fees alone are a massive burden, and many companies are too slow to realize that they are paying to renew obsolete patents.

The patent system may not be in decline, but it is beginning to get breathless in its need to keep up with the searing pace of technological change. The time might therefore be right to step back and consider changes, or parallel systems, to allow necessary protection for fast-moving, if short-lived, technological advantage.


1. World Intellectual Prop. Org., World Intellectual Property Indicators 2017, at 29 (2017), http://www.wipo.int/edocs/pubdocs/en/wipo_pub_941_2017-chapter2.pdf.

2. Statistics: European Patent Applications 2008–2017 per Field of Technology, Eur. Pat. Off. (Jan. 22, 2018), https://www.epo.org/about-us/annual-reports-statistics/statistics.html#filings.

3. China Nat’l Intellectual Prop. Admin. [CNIPA], Patent Statistics Yearbook 2017 (2018).

4. The contracting states to the EPC are: the current member states of the EU (the EU 28); the European Economic Area (EEA) states (Iceland, Liechtenstein, and Norway); and also Albania, Macedonia, Monaco, San Marino, Serbia, Switzerland, and Turkey. Accordingly, the full list of EPC contracting states is: Albania, Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Liechtenstein, Lithuania, Luxembourg, Macedonia, Malta, Monaco, Netherlands, Norway, Poland, Portugal, Romania, San Marino, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, and United Kingdom. In addition, Bosnia and Herzegovina and Montenegro are extension states, and Moldova and Morocco are validation states.

5. Antanina Garanasvili et al., Patenting Strategies in the European Patent System (Dec. 11, 2017) (unpublished manuscript), http://www.oecd.org/site/stipatents/programme/ipsdm-2018-3-1-harhoff.pdf

6. Data as of October 1, 2018, from Darts-ip (https://www.darts-ip.com/).

7. Data as of October 1, 2018, from Darts-ip.

8. Data as of November 1, 2018, held by Gowling WLG (UK) LLP.

9. See, e.g., Warner-Lambert Co. LLC v. Sandoz GmbH [2015] EWHC 3153 (Pat), [2015] EWHC 2924 (Pat), [2015] EWHC 2919 (Pat).

10. 35 U.S.C. § 284.

11. Halo Elecs., Inc. v. Pulse Elecs., Inc., 136 S. Ct. 1923, 1931 (2016).

12. eBay Inc. v. MercExchange, L.L.C., 547 U.S. 388 (2006).

13. Nichia Corp. v. Everlight Ams., Inc., 855 F.3d 1328 (Fed. Cir. 2017).

14. Council Directive 2004/48/EC, 2004 O.J. (L 195) 16.

15. Id. at art. 13.

16. See, e.g., Nikon Corp. v. Carl Zeiss SMT GmbH, LG Mannheim Mar. 2018, 7 O 79/17 (Ger.); Edwards Lifesciences v. Bos. Sci. Corp., LG Düsseldorf Oct. 23, 2018, 4c O 53/18 (Ger.); HTC Corp. v. Nokia Corp., LG Mannheim 2013 (Ger.); IPCom v. Vodafone, LG Düsseldorf Apr. 24, 2012, 4b O 274/10 (Ger.).

17. Unwired Planet Int’l Ltd. v. Huawei Techs. Co. Ltd. [2018] EWCA Civ 2344. In contrast, in Germany, legal tests for the grant of injunctive relief following a finding of infringement of an SEP are more focused on the procedural steps preceding the requests for relief.

18. Conversant Wireless Licensing SARL v. Huawei Techs. Co. Ltd. [2018] EWHC 2549 (Ch).

19. Edwards LifeSciences LLC v. Bos. Sci. Scimed Inc. [2018] EWHC 1256 (Pat). In contrast, in parallel litigation in Germany in 2017, injunctive relief was granted.

20. Regeneron Pharm. Inc. v. Kymab Ltd. [2018] EWCA Civ 1186.

21. Warner-Lambert Co. LLC v. Generics (UK) Ltd. [2016] EWCA Civ 1006; Warner-Lambert Co. LLC v. Actavis Grp. PTC EHF [2015] EWCA Civ 556.

22. Iwncomm v. Sony (Beijing Mun. High People’s Ct. Mar. 28, 2018).


Gordon Harris is a partner at Gowling WLG in London.

Alexandra Brodie is a partner at Gowling WLG.

Jamie Rowlands is a partner at Gowling WLG.