Dietmar Harhoff Institute for Innovation Research, Technology Management and Entrepreneurship (INNO-tec)

Transcription

1 The Social Value of Patent Disclosure Dietmar Harhoff Institute for Innovation Research, Technology Management and Entrepreneurship (INNO-tec) Ludwig-Maximilians-Universität (LMU) München Oxford University St. Peter s College Oxford Intellectual Property Invited Speaker Seminar Series May 26, 2011

2 Agenda Economic Rationales for Patent Systems The Disclosure Debate The Inno S&T Project Measurement of Disclosure EffectsDescriptive Statistics Estimating the Time Savings Distribution Empirical Results Conclusion Dietmar Harhoff - The Social Value of Patent Disclosure 2 Prof. Dietma ar Harhoff, Ph.D D. 2011

3 Economic Rationales for Patent Systems incentive effects due to exclusion rights exclusion rights enhance ex post quasi-rents, R&D and invention efforts are encouraged (Nordhaus 1969) disclosure effects subsequent inventions are facilitated by access to patent literature (Roin 2007) safeguarding transactions, since patents reduce the need for secrecy (Merges 2005) signaling to investors in the presence of asymmetric information patents facilitate acquisition of venture capital (and thus entry) (Harhoff/Haeussler/Mueller 2009) support division of labor by allowing specialists for invention and design to emerge (vertical fragmentation) and by creating markets for technology (Hall and Ziedonis 2001) Dietmar Harhoff - The Social Value of Patent Disclosure 3 Prof. Dietma ar Harhoff, Ph.D D. 2011

4 The Disclosure Debate(1/4) (/) Many scholars and courts see patent disclosure as an important objective of patent systems. Machlup (1958) argues that a patent serves to disseminate technological information, and that this accelerates the growth of productivity in the economy. Eldred vs. Ashcroft (2003): the ( ) twin purposes of the patent system are encouraging new inventions and adding knowledge to the public domain. Towards that objective, patents are required to provide full disclosure in most national and regional patent systems. In Europe, the suggestion of insufficient disclosure is a frequently used and successful attack in oppsoition proceedings. Dietmar Harhoff - The Social Value of Patent Disclosure 4 Prof. Dietma ar Harhoff, Ph.D D. 2011

5 The Disclosure Debate (2/4) (/) A number of legal scholars (e.g., Roin (2009), Fromer (2007) and Seymour (2009)) claim that the system is not working well in the US and that that is due for systematic reasons, inter alia: Disclosure would be important for inventions that are hard to reverse-engineer but those are often protected by secrecy. Patents are opaque, hard to read and to extract information from. Other sources of technical information are easier to use and more suited for search There may be legal l risks in the US in using patents as a source of information - willful infringement rules leading to willful ignorance where ignorance is bliss Roin (2009, 2022f.). Dietmar Harhoff - The Social Value of Patent Disclosure 5 Prof. Dietma ar Harhoff, Ph.D D. 2011

6 The Disclosure Debate (3/4) (/) What do economists do in this debate? The more recent empirical i literature in economics: surveys of R&D managers (e.g. Yale Survey or CIS) show that patents are not used intensively, but there is heterogeity across industries/technical fields. In regressions ess s using relevance eva of patents ts as sources of information as determinants of inventive output or value, these are these without any significant impact (e.g. Arora et al. 2007). However, this form of evidence is very weak. What to make of one Likert-scale point in economic terms? Measurement problems? Do R&D managers really know? Dietmar Harhoff - The Social Value of Patent Disclosure 6 Prof. Dietma ar Harhoff, Ph.D D. 2011

7 The Disclosure Debate (4/4) (/) Summing up The empirical evidence in this debate is extremely thin. Most legal analysis is based on single cases and court statements. The arguments suggesting systematic failure are plausible, but there is virtually no empirical evidence. Economic studies have produced rather soft evidence so far and rely on assessments of managers who may not be deeply involved in the invention processes. Different approach here: ask inventors for quantitative data. Big question: is patent disclosure a contributor to welfare or just an overrated story? Dietmar Harhoff - The Social Value of Patent Disclosure 7 Prof. Dietma ar Harhoff, Ph.D D. 2011

8 Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen. Das Bild kann nicht angezeigt werden. Dieser Computer verfügt möglicherweise über zu wenig Arbeitsspeicher, um das Bild zu öffnen, oder das Bild ist beschädigt. Starten Sie den Computer neu, und öffnen Sie dann erneut die Datei. Wenn weiterhin das rote x angezeigt wird, müssen Sie das Bild möglicherweise löschen und dann erneut einfügen. InnoS&T project EU FP7 project: Innovative S&T indicators combining patent data and surveys: Empirical models and policy analyses a (InnoS&T, Grant Agreement No Collaborative project. Small or medium-scale focused research project) Duration: 36 months 01/04/ /03/2011 Consortium: Bocconi University (coordinator), LMU-Munich, KULeuven and IESE. Slide 8 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

9 Aims of the InnoS&T project 1.develop and collect novel and systematic science and technology indicators 2.develop empirical models that can contribute to improve policies on the following topics: The economic use of the patents: unused patents and strategic patenting; licensing, entrepreneurship Science-industry linkages and innovation performance Gender, education and mobility of inventors The economic value of patents Slide 9 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

10 Aims of the InnoS&T project: S&T Indicators Extensive survey data collections on patent inventors, invention processes, use and value aueof patents: ts: PatVal-EU a II, PatVal-US a and PatVal-JP. a Collection and matching of complementary indicators on patents, inventors, companies, regions, technological fields responsible partners: Bocconi, LMU Creation of novel indicators for industry-science links based on patent citations to science for all OECD countries (over time, across industries, by firms, by universities and by firms to universities and public research institutions) responsible partners: KULeuven, IESE Slide 10 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

11 PatVal-EU II, PatVal-US and PatVal-JP surveys

12 Patval II: Sample Data Source: EPASYS 04/2008, PATSTAT 04/2008 Patents: EP patent applications, priority dates : 327,300 EP patent applications 20 European countries and Israel: the 15 most active European countries in terms of number of granted patents + New Member States and other European countries with at least 50 granted patents in (AT, BE, CH, CZ, DE, DK, ES, FI, FR, GB, GR, HU, IE, IL, IT, LU, NL, NO, PL, SE, SI) USA Japan Slide 12 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

13 Patval II: Sample Questionnaire to inventors: random sample random selection of one inventor for patents with multiple inventors Inventors of multiple patents: one patent of these inventors was selected at random Slide 13 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

14 Composition of the Sample sample size DE 23,032 SE 2,221 FR 8,928 CZ 269 UK 6,018 GR 209 AT 1,528 HU 335 BE 1,539 IE 595 CH 3,068 LU 229 DK 1,108 NO 818 ES 1,318 PL 285 FI 1,358 SI 214 IT 4,875 Europe 62,550 IL 1,241 USA 45,861 NL 3,362 JP 25,000 Total 133,411 Slide 14 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

15 Patval II: Questionnaire Online questionnaire: Languages: English, German, Italian, French, Spanish, Dutch, Hungarian, Czech, Slovenian, Polish Questions to inventors on: Inventors education, age, gender, social setting, employment, mobility, inventions and scientific publications Inventions process: employer (size, age, R&D organization), formal and informal collaborations, sources of knowledge, timing, resources Inventors motivations and rewards: compensation, non monetary rewards Use and value of patents: use of patent for commercial applications, licensing, crosslicensing, foundation of new company Slide 15 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

16 Address Quality - Results Reliable address information: FR, UK, DE, AT, BE, CH, DK, NL, and NO overall share of wrong addresses: ~ 20-30% CZ, ES, FI, GR, HU, IE, IL, IT, LU, PL, SE, SI: online version of the white pages does not provide information small number of patent applications i per country made a search for more recent addresses in the epoline database impossible Corrected response rate (20% wrong addresses): 26.5% Slide 16 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

17 Full scale survey in Europe & Timing of the Responses. PatVal-EU II Dispatch of the invitation letter: Nov. 25, 2009 Dispatch of the reminder letter: Jan. 22, Reminder (2nd cover letter) responses (total) 400 Number of November 29 November 01 December 03 December 05 December 07 December 09 December 11 December 13 December 15 December 17 December 19 December 21 December 23 December 25 December 27 December 29 December 31 December January January January January January January January January January January January January January January January 01 February 03 February 05 February 07 February Slide 17 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

18 Sampling Procedure & Collaboration in the US (a) (b) (c) (d) (e) (f) Random sample of about 50,000 EP applications with oldest priority dates between beginning of 2003 and end of 2005 that contained US inventors; Selection of the oldest US priority for each patent application; Removal of all inventors who do not have a US address; Random selection of one of the remaining inventors per patent; Drop of duplicate inventors (each inventor should only receive one cover letter); Target sample of approximately 45,000 unique inventors on as many unique patents. collaboration with the Massachusetts Institute of Technology (MIT): use of the logo on the letters and envelopes + signature of Prof. Eric von Hippel Slide 18 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

19 Sampling Procedure & Collaboration in Japan (a) (b) (c) (d) (e) (f) Random sample of about 30,000 EP applications with oldest priority dates between beginning of 2003 and end of 2005 that contained Japanese inventors; Selection of the oldest JP priority for each patent application; Removal of all inventors who do not have a JP address; Random selection of one of the remaining inventors per patent; Drop of duplicate inventors (each inventor should only receive one cover letter); Target sample of approximately 25,000 unique inventors on as many unique patents. collaboration with RIETI (Research Institute of Economy, Trade & Industry, Tokyo, Japan): use of the logo on the letters and envelopes + support from Prof. Sadao Nagaoka Slide 19 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

20 RESPONSE RATE (as of February 9, 2011) Overall response rate: 16.4% (20,314 answers). 11.9% complete answers / 4.5% incomplete answers Highest response rates: in Italy, the Czech Republic, and Slovenia. Lowest response rates: US, but reminder postcards have not yet been sent. 40% 35% 30% 25% 20% 15% 16.4% 10% 5% 0% Slide 20 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

21 Response rates AT % 45% % BE % 4.4 % 22.4% CH % 5.5 % 22.1% CZ % 12.7 % 34.7% DE % 52% % DK % 4.1 % 17.4% ES % 6.5 % 20.9% FI % 3.5 % 16.2% FR % 5.2 % 21.1% GB % 3.4 % 14.5% GR % 10.3 % 25.3% HU % 8.6 % 27.2% IE % 4.9 % 19.3% IL % 6.7 % 15.8% IT % 8.4 % 32.4% JP % 10.4 % 25.1% LU % 5.4 % 23.9% NL % 3.9 % 18.0% NO % 2.5 % 14.8% PL % 10.1 % 28.8% SE % 4.0 % 18.1% SI % 47% % US % 1.5 % 6.2% Total % 4.5 % 16.4% Response rates Country Total Responses Complete Incomplete EU and Israel: Overall response rate (as of Dec 4, 2010): 21.6% (complete answers: 16.4%; Incomplete answers: 5.2%) Response rates does not include neutral non-responses : returned letters: 1,244 inventor deceased or ill: 37 no more letters: 48 Total: 1,329 neutral nonresponses (2.14%) US and Japan: reminders and computation of neutral non responses in progress Slide 21 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

22 Results from Patval EU II, Patval-JP, Patval-US Composition of the Patval datasets Technology areas Country Employment position of the invention Employer p y organization Firm size, Age, Ownership by foreign parent company New PatVal II questions on the invention process Origins of the inventions, Working autonomy In other sessions of this conference: Use of the patent (commercial use, licensing, sale, new ventures); Inventors education, gender and mobility, value of the patents Slide 22 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

23 Sample by technological areas Electrical Engineering Mechanical Engineering Chemistry Instruments Process Engineering Consumer goods, Construction 8.23% 15.90% 14.00% 22.92% 19.96% 18.98% 0% 5% 10% 15% 20% 25% Technological Area Freq. Percent Electr/Energy 1, Audiovisual Telecom 1, IT Semiconductors Optical Analysis/Measurement/ControlTechn 1, MedicalTechn NuclearTechn OrganicChem Polymers Pharmaceuticals/Cosmetics Biotechnology Agric&Foods PetrolChem/materialsChem SurfaceTechn Materials ChemEngineering Matprocessing/Textiles/Paper Handl/Printing Agric&FoodProcess-Machines Environment MachineTools Motors ThermProcesses MechElements Transportation 1, SpaceTech/Weapons p ConsGoods ConstrTechn Total 18, Slide 23 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

24 Employment status at the time of the invention Student; 0.9% Other; 4.6% % selfemployed Country GR 27.8% NO 18.3% IT 17.6% AT 15.5% Unemployed Selfemployed ; ; 0.3% ES 14.9% 8.0% IE 13.4% PL 12.9% SI 12.7% NL 12.2% Employed; 86.2% CZ 11.7% HU 10.7% SE 10.2% CH 9.8% DK 9.6% BE 8.7% Total 8.0% DE 7.9% Employed Self-employed employed Unemployed GB 7.8% US 7.5% Student Other IL 6.4% FI 5.4% B1 - Employment status at the time of the invention (N = 16,959) LU 53% 5.3% FR 2.7% JP 0.9% Slide 24 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

25 Organization/employer at the time of the invention 2.3% 0.4% 1.0% 5.9% 4.1% 86.3% Private firm University or other educational institution Private hospital, research organization or foundation Government research organization Other government organization Other B2 Type of organization/employer the inventor was affiliated with at the time of the invention (N=16,692) Country Firm PRI Other LU 94.7% 0.0% 5.3% JP 94.6% 45% 4.5% 09% 0.9% CH 90.7% 5.8% 3.5% FI 90.5% 6.5% 3.0% DK 90.4% 6.2% 3.4% SE 90.2% 55% 5.5% 42% 4.2% AT 90.1% 4.3% 5.7% DE 89.2% 4.7% 6.1% Total 86.3% 6.8% 6.9% IT 85.8% 8% 77% 7.7% 65% 6.5% BE 84.7% 8.1% 7.2% FR 84.2% 11.8% 4.0% NO 82.5% 7.0% 10.5% GB 82.4% 10.4% 73% 7.3% US 80.6% 4.8% 14.6% HU 79.5% 8.2% 12.3% NL 76.3% 10.2% 13.5% IE 75.3% 20.6% 41% 4.1% PL 73.3% 21.7% 5.0% CZ 73.0% 14.9% 12.2% ES 72.5% 14.4% 13.1% IL 70.0% 0% 18.1% 1% 11.9% SI 67.7% 16.1% 16.1% GR 64.7% 26.5% 8.8% Slide 25 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

26 Number of employees in the organization 45% 40% 40.2% 35% 30% 25% 20% 18.0% 15% 10% 5% 7.8% 3.0% 4.8% 4.4% 7.7% 6.7% 7.5% B3-0% Number of employees working in the organization at the time of the invention (N = 16,124) Slide 26 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

27 Number of employees in the organization, by country 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% >500 B3 aggregate - Number of employees working in the organization at the time of the invention (N = 16,124) Slide 27 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

28 Majority owner of the employer: Foreign Parent Company Country Foreign parent company HU 37.84% LU 37.50% BE 32.20% CZ 30.26% GB 28.04% PL 27.87% AT 27.27% foreign parent SI 25.40% company IE 24.74% 16.34% NL 20.82% ES 20.80% IT 20.23% FR 19.32% NO 18.42% DE 18.11% SE 17.28% Total 16.34% CH 14.87% FI 14.78% IL 13.94% DK 11.24% 83.66% GR 11.11% B2A - Foreign Parent Company (N = 16,844) US 10.37% JP 2.62% Slide 28 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

29 Age of the organization at the time of the invention 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% <2 2-4 >5-10 >10-20 >20 B7A - Age of the organization the inventor was employed with at the time of the invention (N = 11,183) Slide 29 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

30 Organization of the inventive activities leading to the patented invention Number of employees in the organization Percentage done through individual work % % % 08% % % 50% Individual work, 34.4% Team work, 65.6% Total 34.09% 40% % % 30% % 20% % 10% C3 - Organization of the inventive activities leading to the patented invention (N=15,491) 0% Slide 30 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

31 Origin of the idea for the invention Other person in team; 5.9% Idea of team, incl. me; 42.6% Person not in team; 0.9% Other; 1.7% My idea; 48.9% During leisure time - hobbies; 10.4% During leisure time -work at home; 8.9% Other; 3.5% My idea Idea of team, incl. me During Other person in team Person not in team work time; Other 77.2% C35 - Origin of the idea for the invention (N = 14,195) C36 - Information about the origin of the idea for the invention (N = 12,903) Slide 31 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

32 Frequency with which the inventor used to think about the invention outside the working time (0. never; 1. occasionally - 5. very often) 40% 35% 34.0% 30% 25% 20% 15% 16.1% 11.7% 16.2% 18.3% 10% 5% 3.8% 0% D6 - Frequency with which the inventor used to think about the invention outside the working time (N = 13,637) Slide 32 Feb , 2011 InnoS&T Conference, Munich Myriam Mariani / Karin Hoisl

33 Measurement of Disclosure Effects How to operationalize and measure the effects of disclosure? approach taken here: patent literature contains technically relevant information reading patents allows inventors to apply that information to subsequent inventions cost reduction due to knowledge of the patent literature We seek to measure this cost reduction at the level of patent applications (not inventors) within the context of an inventor survey. Dietmar Harhoff - The Social Value of Patent Disclosure 33 Prof. Dietma ar Harhoff, Ph.D D. 2011

34 Measurement of Disclosure Effects Cost Savings from Patent Disclosure Dietmar Harhoff - The Social Value of Patent Disclosure 34 D ar Harhoff, Ph.D Prof. Dietma

35 Measurement of Disclosure Effects Comments on the survey yquestion the tailor-made response items were used in 1,080 out of 7,837 cases a very small number of inventors indicated negative cost savings (i.e., a cost burden from reading patent literature) share of don t know responses fairly large 27.9% of 10,874 observation Dietmar Harhoff - The Social Value of Patent Disclosure 35 Prof. Dietma ar Harhoff, Ph.D D. 2011

36 Descriptive Statistics Importance of Patent Literature Main Technological Area compl. unimportant unimportant Important very important not used Total Electrical Engineering Instruments Chemistry Process Engineering Mechanical Engineering Consumption and Construction Total Dietmar Harhoff - The Social Value of Patent Disclosure 36 Prof. Dietma ar Harhoff, Ph.D D. 2011

37 Descriptive Statistics time saving due to knowledge of patent literature Freq. Percent Cum. 0 2 hours 2, >2 5 hours >5 10 hours >10 20 hours >20 50 hours > hours more than 100 hours not easily quantified, but definitely more than not easily quantified, but definitely less than don't know 3, Total 10, Dietmar Harhoff - The Social Value of Patent Disclosure 37 Prof. Dietma ar Harhoff, Ph.D D. 2011

38 Estimating the Time Savings Distribution ordered probit estimator with known threshold (standard in some statistics packages, e.g. intreg in Stata) using logarithms of threshold h values to delineate intervals covariates: initially only priority year and technical area (30 groups, OST-INPI classification, 6 main groups, country dummies) coefficents are identified via threshold information predicted values are then used to compute moments (alternative route: compute moments analytically from parameters of the log-normal distribution) Dietmar Harhoff - The Social Value of Patent Disclosure 38 Prof. Dietma ar Harhoff, Ph.D D. 2011

39 Estimating the Time Savings Distribution likelihood function (Source: Stata Manual, intreg) Dietmar Harhoff - The Social Value of Patent Disclosure 39 D ar Harhoff, Ph.D Prof. Dietma

40 Empirical Results Moments of the Time Savings Distribution by Main Technological Area (conditional on using patent literature) Main Technological Area p10 p50 mean p90 N Electrical Engineering ,661 Instruments ,285 Chemistry ,529 Process Engineering ,138 Mechanical Engineering ,604 Consumption and Construction Total ,834 Dietmar Harhoff - The Social Value of Patent Disclosure 40 Prof. Dietma ar Harhoff, Ph.D D. 2011

41 Empirical Results Moments of the Time Savings Distribution by Main Technological larea (including cases with patents not used as sources of information) Main Technological Area p10 p50 mean p90 N Electrical Engineering ,294 Instruments ,585 Chemistry ,751 Process Engineering ,430 Mechanical Engineering ,027 Consumption and Construction Total ,888 Dietmar Harhoff - The Social Value of Patent Disclosure 41 Prof. Dietma ar Harhoff, Ph.D D. 2011

42 Empirical Results Technological Area p10 p50 mean p90 N Electrotechnical and Energy Audiovisual Telecom IT Semiconductors Optical Analysis, Measurement, Control Technology Moments of the Time Savings Distribution ib ti IT by Technological Area (including cases with patents not used as sources of information) MedicalTechnology Nuclear Technology Organic Chemicals Polymers Pharmaceuticals, Cosmetics Biotechnology Agriculture, Foods Petrochemical, Materials Chemistry Surface Technology Materials Chemical Engineering Materals Processing, Textiles, Paper Handling, Printing Agricultural and Food Processing Machines Environment MachineTools Motors Thermical Processes Mechancial Elements Transportation Space, Weapons ConsGoods Construction Technology Dietmar Harhoff - The Social Value of Patent Total Disclosure , Prof. Dietma ar Harhoff, Ph.D D. 2011

43 Empirical Results ElecEng Instruments Chemistry Density ProcEng MechEng ConsConstr log(time saved due to disclosure) Graphs by main technological area (ass. with IPC) Density kdensity lnts Dietmar Harhoff - The Social Value of Patent Disclosure 43 Prof. Dietma ar Harhoff, Ph.D D. 2011

44 Conclusions (1/3) (/) Disclosure effects appear to be relatively small, but with some ht heterogeneity across technical hi lfild fields. Median Mdi (p90) (90) values 1.2 (2.7) hours in telecommunication 27.6 (99.8) hours in organic chemicals High (non-trivial) correlation between appropriation and disclosure effects. In areas in which important inventions are patented, inventors read patents and learn from them. Turned around: a (counter-factual) limitation of patent rights to areas like pharmaceuticals and chemicals is unlikely to cause cost increases for follow-up inventions. Dietmar Harhoff - The Social Value of Patent Disclosure 44 Prof. Dietma ar Harhoff, Ph.D D. 2011

45 Conclusions (2/3) (/) Very little evidence in favor of the positive disclosure story adopted by some courts, in particular the Supreme Court. Learning from past patents is a highly overrated story, at least as regards the facts. Normatively speaking, would we want to strengthen disclosure? Should there be more stringent requirements on disclosure? plain language (Bessen and Meurer 2009) invigorated patent disclosure (Fromer 2009) Agreement here there may be high costs in shifting towards radically more stringent standards. Moreover, the effects of that are highly uncertein. Dietmar Harhoff - The Social Value of Patent Disclosure 45 Prof. Dietma ar Harhoff, Ph.D D. 2011

46 Conclusions (3/3) (/) Importants Caveat - Does disclosure have significant other effects? avoiding duplication (coordination)? promoting efficient i stopping of R&D just the spark triggering inventions, but no cost reduction Are there better ways of measuring the impact of disclosure? dscosue? Many problems and weaknesses, but this is the only y( (non-qualitative) measurement attempt to date. Dietmar Harhoff - The Social Value of Patent Disclosure 46 Prof. Dietma ar Harhoff, Ph.D D. 2011

47 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 München Tel. +49 (0) Fax +49 (0) tec Dietmar Harhoff - The Social Value of Patent Disclosure 47 Prof. Dietma ar Harhoff, Ph.D D. 2011