Patent Portfolio Constructionism and Strategic Patenting Dietmar Harhoff Institute for Innovation Research, Technology Management and Entrepreneurship (INNO-tec) Ludwig-Maximilians-Universität (LMU) München Presentation Prepared for the TILEC Conference on Patent Reforms March 26/27, 2009 Amsterdam Overview Articles in the Background What I heard this morning The Patent System as a Selection Instrument Motivation Rembrandts and more Patent Constructionism Defined Examples Trends Policy Reforms Further Recommendations Dietmar Harhoff - Patent Portfolio Constructionism 2 1
Publications in the Background Wissenschaftlicher Beirat des Bundesministeriums für Wirtschaft und Technologie (2006). Innovation und Patentschutz. Harhoff, D. (2009). Economic Cost-Benefit Analysis of a Unified and Integrated European Patent Litigation System, Final Report, Tender No. MARKT/2008/06/D Von Gaevenitz et al. (2008). The strategic use of patents and its implications for enterprise and competition policies, Final Report, Tender EU ENTR 05 82 Harhoff, D., K. Hoisl und B. van Pottelsberge de la Pottherie (2008). Languages, Fees and the Regional Scope of Patenting in Europe, CEPR Discussion Paper, Centre for Economic Policy Research, London. Dietmar Harhoff - Patent Portfolio Constructionism 3 What I heard this morning There is no problem This is a system for legal people. not for technical people nor for economists ( listening to them for half a day ) More patents are better. Injunctions are good. Is there any problem? YES Where (who) is the problem? Dietmar Harhoff - Patent Portfolio Constructionism 4 2
Motivation Rembrandts and more The Young Patent Attorney s Introduction to the Real World ( ) On my first day working as a patent attorney, the senior partner came in and talked to me. He said: Look, any idiot can file applications for real inventions. True genius will only show when you take care of the rest of the crap. - anonymous - Dietmar Harhoff - Patent Portfolio Constructionism 5 Motivation Rembrandts and more 1800 1600 1400 1200 Patent Value Distribution 1,690 1,601 1,348 1,198 Frequencies 1000 800 600 400 200 0 603 < 30K 30-100K 100-300K Dietmar Harhoff - Patent Portfolio Constructionism 6 743 287 155 300-1M 1-3M 3-10M 10-30M 30-100M 100-300M Value Classes (7752 obs.) Rembrandts? 59 68 > 300M 3
Patent Systems as Selection Instruments fees attorneys translation cost-benefit selection 50% examination Exemplary values for EPO opposition litigation R&D other applications grants (some) public policy issues pricing??% 33% stringency subject matter rules of the game 2 (4)% duration costs/benefits rules of the game 0.15% harmonization costs/benefits balance Dietmar Harhoff - Patent Portfolio Constructionism 7 Patent Constructionism Defined For the purpose of this presentation strategies and tactics used by patent applicants to construct patent portfolios by constructing overlapping, multiple filings with high similarity from smaller building blocks (claims, first filings) or by recombination of smaller bulding blocks (claims, first filings) In the extreme: single patents become irrelevant the portfolio is redundant. Often goes hand in hand with other strategies seeking to increase uncertainty inflation of claims, keeping applications in play, Dietmar Harhoff - Patent Portfolio Constructionism 8 4
Patent Constructionism Defined Some Caveats no behavior displayed here is illegal nonetheless, it might be harmful to competition and innovation, in particular by increasing uncertainty patent-holders competing against each other have little choice to adopt such behaviors (even if they are harmful to the overall system) code of conduct of patent representatives appears possible, but has been discouraged by some applicants ultimate bastion of quality control : the patent offices last but last bastion: post-grant reviews and courts Dietmar Harhoff - Patent Portfolio Constructionism 9 Examples from a series of 7 fililngs (ANGIOTECH) WO 2005/046747 A2-1,738 claims WO 2005/046746 A2-10,247 claims WO 2005/051444 A2-19,368 claims filings with excessive numbers of claims and other rule-breaking behaviors no real sanction available impact: on uncertainty, transaction costs Dietmar Harhoff - Patent Portfolio Constructionism 10 5
Examples Based on the 7 WO/PCT filings, the applicant has filed more than 50 U.S. applications Claims were typically amended to a much lower number (about 100 for each of the applications) note: the USPTO charges claims fees. In the case of publication number US2005/0182468, the applicant was initially requested to pay further fees in the order of 1.3 million US$ (and reduced the number of claims to from 13,305 to fewer than 70). The amount reflects fees for 9,999 claims. The respective application originally contained 13,305 claims. Dietmar Harhoff - Patent Portfolio Constructionism 11 Examples Priority Application Patent Divisional Patent Divisional Patent EP19940903359 EP0673578 B1 EP19980105670 EP19940903360 EP0673579 B1 EP19990100566 EP19970112080 EP0822718 B1 EP20010129011 EP19980103750 EP0862328 B1 EP19940903361 EP0673580 B1 EP19990110233 EP0946060 B1 EP20040001389 EP1432248 B1 EP20040001389 US19920991074 EP19940903362 EP0673581 B1 EP19990107757 EP19980105647 EP0856993 B1 EP20010113921 EP19940903407 EP0673582 B1 EP19980121389 EP0909095 B1 EP19980100155 EP0849948 B1 EP19980114676 EP19990100375 EP0912058 B1 EP19940904392 EP0673583 B1 EP19990100376 EP19990100377 EP0920206 B1 EP19990100378 EP0920207 B1 EP19990100570 EP0910218 B1 EP19940904814 EP0674824 B1 EP19980100142 EP0852442 B1 Dietmar Harhoff - Patent Portfolio Constructionism 12 6
Examples Priority application US19920991074 ( cable set-top box, 91 claims, 183 pages description) was used to file 7 (very similar) EPO applications. An additional 16 divisionals were filed based on the first 7 applications. Another 3 (second generation) divisionals were filed on the basis of the first round of divisionals. Summing up 26 applications of which (to date) 18 were granted. Exceptional? No, extreme within a comprehensive trend Dietmar Harhoff - Patent Portfolio Constructionism 13 Examples publication number EP1286749, applicant: Genentech Inc. title: Secreted and Transmembrane Polypeptides and Nucleic Acids Encoding the Same characteristics: 22 claims, 11 inventors, 30 priorities (mostly US and WO) 16 divisional applications EP1683864, EP1657253, EP1666597, EP1666490 EP1666596, EP1686174, EP1657254, EP1666491 EP1666492, EP1659177, EP1666493, EP1666497 EP1666594, EP1657251, EP1700867, EP1702928 Dietmar Harhoff - Patent Portfolio Constructionism 14 7
Examples Patents in Cosmetics EP1707181, EP1707182, EP1707183, EP1707184, EP1707190 priorities: FR20050050835 20050331 - FR20050050838 20050331 - FR20050050841 20050331 - FR20050050837 20050331 - FR20050050842 20050331 for all of these EPO filings: application date: March 28, 2003 publication date: October 4, 2006 applicant: major player in the cosmetics industry same inventor: Legrand Frederic between 24 and 28 claims (industry average 12.2 claims) same examiner Dietmar Harhoff - Patent Portfolio Constructionism 15 Examples Patents in Cosmetics Title of EP1707181 Dye composition with a reduced content of starting materials, and process for dyeing keratin fibres using the same Title of EP1707182 Dye composition comprising a fatty acid and process for dyeing keratin fibres using the same Title of EP1707183 Dye composition comprising a non-ionic associate polymer, process for dyeing keratin fibres using the same Title of EP1707184 Dye composition with a reduced content of starting materials, process for dyeing keratin fibres using the same and device therefore Title of EP1707190 Dye composition comprising a hydrophobically modified nonionic cellulose and method for dyeing keratin fibres using it Dietmar Harhoff - Patent Portfolio Constructionism 16 8
Trends increasing number of claims increasing share of X-references in applications declining opposition rate larger patent families (by any definition, for within-ep equivalents from 1.05 in 1978 to 1.16 in 2000) soaring number of divisional filings increasing backlogs and times to first office action - global patent warming applications (not grants) have strategic effects (increasingly so) Filings are up, complexity is up, quality is down, the grant rate is constant (or has been till recently). Dietmar Harhoff - Patent Portfolio Constructionism 17 Trends Average Number of Claims 25 20 Claims 15 10 5 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 Application Year US-Priority US-Priority/PCT JP-Priority JP-Priority/PCT DE-Priority DE-Priority/PCT Dietmar Harhoff - Patent Portfolio Constructionism 18 9
Trends Average Share of X-Type References 30 25 Share (%) 20 15 10 5 0 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 Application Year US-Priority JP-Priority DE-Priority Dietmar Harhoff - Patent Portfolio Constructionism 19 Trends Divisionals 0 2,000 4,000 6,000 8,000 Divisionals by Publication Year 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Source: March 2007 EPASYS & April 2007 PATSTAT Dietmar Harhoff - Patent Portfolio Constructionism 20 10
Trends Opposition Frequency by Grant Year Opposition Frequency (%) 14 12 10 8 6 4 2 0 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 Grant Year Electrical Engineering Instruments Chemistry Process Engineering Mechanical Engineering Dietmar Harhoff - Patent Portfolio Constructionism 21 Policy Reforms London Protocol leads to significant reductions in patenting costs will lead to an increase in patent validations by (estimated) 29% (Harhoff, Hoisl, van Pottelsberghe 2009, CEPR DP) Patent Litigation see my recent report to the Commission preference for a low-cost, unified system with strong revocation options change of metaphor required patent litigation (largely) as innocuous car accidents Dietmar Harhoff - Patent Portfolio Constructionism 22 11
Further Recommendations Governance of the EPO remove implicit incentives for EPC delegates in the adminstrative council to vote in favor of more grants raise requirements for novelty and inventive step lower the grant rate Incentives for Examiners remove decision-making biases at the examiner level - make examiners indifferent between grant and refusal give examiners more leeway to reject badly drafted patents Fee Structure sanction strategic patent design (claims, complexity, intentionally bad quality ) do not make patents more affordable across the board Dietmar Harhoff - Patent Portfolio Constructionism 23 Further Recommendations Allow for opposition and litigation to attack similar patents simultaneously. patents sharing priority filings patents sharing claims and elements of the description Private incentives are partly disabled in patent thickets. Introduce a function that takes care of the public good (similar to: representatives of the public interest in German courts). Wagner/Pachomovsky: set prices for similarity and for filing strategies that emphasize patent constructionism (e.g. require higher fees from applicants with large portfolios). Dietmar Harhoff - Patent Portfolio Constructionism 24 12
Wrap-up The question whether patent protection is good or not good may be outdated. Big question: what is the optimal design for a patent system from an economics perspective (=to maximize welfare). This will require more attention to institutional detail, both in theoretical and empirical work. As usual with organization design it will be science and art. An early insight by Nordhaus (1972, p. 430f): The best way to prevent abuse is to ensure that trivial inventions do not receive patents. Dietmar Harhoff - Patent Portfolio Constructionism 25 Thank you for your attention! Prof. Dietmar Harhoff, Ph.D. Institute for Innovation Research, Technology Management and Entrepreneurship Ludwig-Maximilians-Universität München (LMU) Munich School of Management Kaulbachstr. 45 D-80539 München Tel. +49 (0)89-2180-2239 Fax +49 (0)89-2180-6284 harhoff@bwl.uni-muenchen.de http://www.inno-tec.de Dietmar Harhoff - Patent Portfolio Constructionism 26 13