Difference between revisions of "Untangling the Economics of Patent Thickets"

From edegan.com
Jump to navigation Jump to search
 
(16 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
{{AcademicPaper
 
{{AcademicPaper
|Has title=Patent Thicket Literature Review
+
|Has title=Untangling the Economics of Patent Thickets
 
|Has author=Ed Egan, David Teece
 
|Has author=Ed Egan, David Teece
 
|Has RAs=Lauren Bass,
 
|Has RAs=Lauren Bass,
 
|Has paper status=Working paper
 
|Has paper status=Working paper
 
}}
 
}}
==The Paper==
 
  
Latest version is:
+
==Latest Progress==
 +
The PDF-to-Text converter, the key terms finder, and phrase extraction scripts all work well and have been tested. The Google Scholar Web Crawler is being worked on by Christy Warden, current status unknown.
  
\coauthoredprojects\Egan and Teece\McN-PatentThicket-Egan-092215.pdf
+
192 text versions of papers have been assembled in "Candidate Papers by LB." These were used as a test set for developing an analysis protocol for the entire paper. An analysis protocol has been developed that involves taking the counts of each term by topic per paper to assign each paper 1-4 topics. The process also counts the number of key terms in each paper for relevance, the number of modern terms, the year of the paper, the number of cost terms, the area of the paper, and the mentions of particular authors. All of this analysis is then dumped into artifacts. The analysis protocol is currently hardcoded. Once a set list of terms has been established, a soft-coded version should be able to be developed (in progress).
  
This file is posted at http://www.bakerinstitute.org/research/untangling-patent-thicket-literature
+
Larger test set desired to be analyzed and soft-coded on.  
  
==Definitions==
+
==The Paper==
  
===Core Terms===
+
The latest version is:
  
====Saturated Invention Spaces====
+
\coauthoredprojects\Egan and Teece\McN-PatentThicket-Egan-092215.pdf
  
Description:
+
This file is posted at http://www.bakerinstitute.org/research/untangling-patent-thicket-literature
The earliest definition of a patent thicket that we found is in Teece (1998), who points out a simple issue
 
with patents. We refer to this as ‘saturated invention spaces’ and describe it as when a single firm,
 
or a small number of firms, successfully patents an entire technological area. (p.15)
 
 
 
Market Failure: Imperfect competition: a small number of firms hold all possible patent rights in an area
 
 
 
Look for:
 
Terms like “cluster”, “related”, “adjacent”, or “coherent group” to describe either patents or inventions.
 
However, we also included several definitions that described patents as “minor” or “marginal” provided
 
that there was no suggestion of bad faith
 
 
 
B) '''Diversely-held Complimentary Inputs (DHCI):''' 1) products require complementary patented inputs; 2) these inputs are diversely-held (i.e. held by N patent-holders); and 3) patent-holders set their license prices independently. (Shapiro, pg.17)
 
* '''''Coordination: patent holders determine licensing rates independently'''''
 
 
 
“Diversely-held” and “complementary inputs,” or clear synonyms like “dispersed” or “fragmented” pertaining to ownership and “Cournot problem”, “multiple marginalization”, etc. in the context of licensing of patents held by multiple parties
 
 
 
C) '''Overlapping Patents:'''
 
#The second most common foundation for a patent thicket issue relies on overlapping patents. Patent overlaps can be horizontal (i.e. though patents that are largely adjacent to one another) or vertical (i.e. when  patents are related through cumulative innovation). Horizontal overlaps arise because patent rights are imperfectly defined property rights. (p.20)
 
#Refinement patents and research tool patents could result in vertically overlapping patent rights (or at least contractual rights). (p.21)
 
* '''''Imperfectly defined property rights: wasteful duplication of licensing expenditure'''''
 
 
 
Any sense of patent overlaps including “overlapping claims” or “similar claims”, and/or uncertainty over whether multiple patents may simultaneously be infringed
 
 
 
D) '''Gaming the Patent System:''' There are information asymmetries between a patent applicant and the patent office and a patent applicant may take some inappropriate action, for example by applying for a patent that is not novel or is obvious. Such an inappropriate action can impose costs on the patent office and may generate negative externalities, imposing costs on genuine innovators. We refer to this situation as ‘gaming the patent system’. (p.21)
 
*'''''Moral hazard: patent applicants take hidden action and impose costs on genuine innovators'''''
 
 
 
Primarily descriptions of patents as “questionable” “dubious” “bad”, “likely invalid” and “junk,” and phrases congruent with “impeding genuine innovators” or “rent-seeking activities”, as well as other indications of bad faith
 
 
 
====Modern Terms====
 
 
 
E) '''Transaction Costs:''' All patent-based interactions, from application to licensing to litigation, are subject to transaction and search costs.  Patent applications are subject to transaction costs in the form of prosecution costs and renewal (‘maintenance’) fees.  These costs and fees are supposed to de-incentivize low value patents. However, they may also de-incentivize invention by small firms and individuals. (p.13)
 
 
 
F) '''Probabilistic Patents:''' Lemley and Shapiro (2005, 2006) emphasize that patents are ‘probabilistic’. They suggest that “there is no way to determine with certainty whether the patent is valid and infringed without litigating to judgment." Enforcement of patent rights - observing and redressing infringement of patent claims - is therefore both costly and uncertainty. (p.11)
 
 
 
G) '''Unspecified/Extended Use:''' Kiley (1992) claims that applying for a patent based on an inventive step that does not have a clearly articulated stand-alone commercial value creates economic inefficiencies. But often it can be difficult to foresee which particular use of an invention will be profitable and which won’t. (p.8)
 
 
 
H) '''Search Costs:''' Searching for relevant patents may impose material costs. Wang (2010) argues that this may be particularly burdensome for new entrants who need to develop suitable search capabilities. (p.13)
 
 
 
I) '''(Patent) Hold-up:''' The Federal Trade Commission (2011) provides the following definition of patent hold-up: “‘Hold-up’ describe[s] a patentee’s ability to extract a higher license fee after an accused infringer has sunk costs into implementing the patented technology than the patentee could have obtained at the time of [the accused infringer’s] design decisions.” However, some recent papers have stressed the obvious diametrically opposite problem. Langus et al. (2013), for example, point out that
 
:''“the licensee may often engage in a reverse hold up”. Reverse patent hold-up, loosely put, is where the accused infringer extracts zero (via infringing) license fee after the patent owner has sunk costs in developing :the patented technology and alleged infringement has taken place. Again, this occurs as patents are not self-enforcing. Implementers can simply use the invention covered by a patent and wait to get sued, using as :many diversionary tactics in the courts as is possible, knowing that it is hard, time-consuming, and expensive for a patentee to get an injunction. The judicial system is far from perfect; the patentee has few remedies :absent a courts intervention.'' (p.11-12)
 
 
 
J) '''Strategic Patents:''' Strategic patenting is sometimes defined as accumulating patents merely to achieve design freedom. These patents can be used as bargaining chips, rather than for their intrinsic value, and as such are largely welfare neutral, except in conjunction with transaction costs and cross- licensing agreements (discussed shortly). Much of the discussion of strategic patents takes place in the context of the patent thicket literature. (p.13)
 
:''“To obtain the rights to infringe patents held by external parties and to improve their leverage in negotiations with other patent owners, these firms amass larger patent portfolios of their own with which to trade.” – :Hall and Ziedonis (2001)''
 
 
 
K) '''Hold-out:''' Hold-out can occur in the context of multilateral bargaining, for example when different products can be made out of different, diversely-held patented technologies. Farrell (2009) explains that when a partial agreement benefits the “nonparticipating (holdout) player”, self-interest and social welfare may not be aligned. (p.12-13)
 
 
 
===Relative Definition Quotes===
 
 
 
Shapiro: "a dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialize new technology" (p.3-4)
 
 
 
Kiley: "Those operating at the beginnings of the road are most insistent on their right to nail down leverage that will remain formidable despite market place rejection of the uses to which they say their inventions may be put.  The frank aim of these early stage workers is to control ultimate applications discovered by others." (p. 8)
 
 
 
 
 
=Measures=
 
 
 
==Types==
 
  
# '''Theory:''' Economic theory papers are largely concerned with the DHCI problem. (p.5)
+
<pdf>File:McN-PatentThicket-Egan-092215.pdf</pdf>
# '''Empirical:''' Papers that drift towards first including overlapping patents along with DHCI, and then begin abandoning the usage of DCHI. (p.5)
 
# '''Discussion:''' It seems that economists favor DHCI and overlaps, cite Shapiro (2001) and do more theoretical work; whereas legal scholars favor overlaps, cite Heller and Eisenberg (1998) and provide discussion. (p.30)
 
  
==Industries==
+
==Codification==
  
# '''Basic Science:''' Basic research, which includes the commercialization of academic research, biotechnology, genetics, nanotechnology, and pharmaceuticals. (p.28)
+
See the [[Patent Thicket Strategic Planning]] page for future steps.  
# '''Complex Science:''' Complex product industries, which includes information and communications technology (aside from software and the internet but specifically including semiconductors), manufacturing, and sewing machines (which in the period of analysis of papers discussing it was a complex cutting-edge product). (p.28)
 
# '''Software / Bus. Method:''' An aggregate of firms focused on software, business methods, and the internet. (p.28)
 
  
==Topics==
+
See the [[PTLR Codification]] page for details
 
 
# '''Effects on Academia:''' Overlapping patents are curiously related to basic science and the effects of patenting on academic research, both of which depend upon cumulative innovation. (p.30)
 
# '''Private Mechanism:''' Private arrangement papers discuss cross-licensing, patent pools, patent clearinghouses, patent collectives, FRAND licensing agreements, patent intermediaries (including NPEs), shared platforms, technology standards, and Standard Setting Organizations (SSOs). (p.27)
 
# '''Industry Commentary:''' There was also a distinct set of publications that engaged in industry commentary; commentary on the nanotechnology and genetics industries were particularly common. (p.27-28)
 
# '''IPR Reform:''' IPR reform papers suggest reforms to the nature of intellectual property rights, examine processes for granting patents at the patent office, and advocate approaches to patent-related transactions for anti-trust authorities. (p.27)
 
# '''Firm Strategy:''' Firm strategy papers provide strategic advice to firms regarding their intellectual property – they discuss the strategic implications of blocking patents, pre-emptive patenting, secrecy, ever-greening, avoiding willful infringement, engaging in Mexican standoffs, and other defensive or offensive patenting behaviors, as well the consequences of doings so on collaboration, industry structure (including entry), and the value of firms. (p.27)
 
# '''Patent Thickets:''' We suggest that one reason why papers implement definitions that are consistent with more than one economic issue is that many of the original definitions of patent thickets were made by analogy. Analogies face the risk of multiple possible interpretations. Shapiro (2001) is frequently quoted in the literature saying that patent thickets are “a dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialize new technology". Shapiro’s analogy is compatible with all four of the most commonly discussed economic issues referred to as patent thickets. (p.3-4)
 
 
 
==Publication Type==
 
 
 
# '''Econ'''
 
# '''Law'''
 
# '''Science'''
 
# '''Policy Report'''
 
 
 
# '''Number of Authors'''
 
# '''Repeat Authors'''
 
 
 
==Current List of Papers==
 
 
 
===Processed papers===
 
  
 +
See the [[PTLR Process]] page for details
  
 
==Test Run==
 
==Test Run==
Line 115: Line 35:
 
Example BibTeX:
 
Example BibTeX:
  
  @article{clarkson2005patent,
+
  @article{BibTeXKeyGoesHere,
   title={Patent Informatics For Patent Thicket Detection: A Network Analytic Approach For Measuring The Density Of Patent Space},
+
   title={},
   author={Clarkson, G.},
+
   author={},
   journal={Academy of Management, Honolulu},
+
   journal={},
   year={2005},
+
   year={},
   abstract={When organizations in technology industries attempt to advance their innovative activities, they may encounter patent thickets, or dense webs of overlapping intellectual property rights owned by different companies that must be hacked through in order to commercialize new technology. Throughout the last 150 years, however, organizations have stumbled into a number of patent thickets and have occasionally responded by constructing patent pools or organizational structures where multiple firms collectively aggregate patent rights into a package for licensing, either among themselves or to any potential licensees irrespective of membership in the pool. Such collaboration among technologically competing firms, however, has often encountered difficulty from an antitrust standpoint, even if the formation of the pool is pro-competitive. Despite all that has been written lamenting the problem of patent thickets, the antitrust regime has never had an objective method of verifying the existence of a patent thicket in a given section of patent space. In response to the lack of such a methodology, this paper proposes a tool to facilitate objectively demonstrating the existence of patent thickets. This paper proposes a thicket identification methodology that uses a network analytic technique to determine if a patent pool is coincident with a patent thicket by comparing the density of the patent pool to the density of the surrounding patent space. This paper then applies the new methodology to two existing patent pools and verifies the existence of underlying patent thickets.... Patent thickets are not a new phenomenon, and when the total number of owners of the conflicting intellectual property rights is small, the response to the patent thicket problem has often been to cross-license (Grindley & Teece 1997; Teece 1998; Teece 2000). When more than two parties are involved, however, the transaction costs of cross-licensing between all of the parties can be prohibitive, and additional economic barriers exist such as hold-ups and double marginalization (Viscusi et al. 2000).},
+
   abstract={},
   discipline={Law, Econ},
+
   discipline={},
   research_type={Measures},
+
   research_type={},
 
   industry={},
 
   industry={},
 
   thicket_stance={},
 
   thicket_stance={},
 
   thicket_stance_extract={},
 
   thicket_stance_extract={},
   thicket_def={#A-T, #B-T, Quotes Shapiro, References Shapiro, References Heller/Eisenberg, Diversely-Held, Transaction Costs, Unspecified Blocking Mechanism, Overlapping Patents, Complementary Inputs, Hold-up, Cummulative Invention},
+
   thicket_def={},
   thicket_def_extract={When organizations in technology industries attempt to advance their innovative activities, they may encounter patent thickets, or dense webs of overlapping intellectual property rights owned by different companies that must be hacked through in order to commercialize new technology... Using Shapiro’s definition of a patent thicket as the starting point, two conditions must be satisfied in order for a collection of patents to be a patent thicket: the collection of patents must be both “dense” and “overlapping”(2000, pg. 120).},   
+
   thicket_def_extract={},   
 
   tags={},
 
   tags={},
   filename={[[Clarkson (2005) - Patent Informatics For Patent Thicket Detection]].pdf}
+
   filename={}
 
  }
 
  }
  
 +
===Files===
  
 +
Pick 5 files at random from:
 +
Dropbox\coauthoredprojects\Egan and Teece\CandidatePapers
  
===Files===
+
List them here:
 +
*Cockburn MacGarvie (2006) - Entry And Patenting In The Software Industry.pdf
 +
*Murray Stern (2007) - Do formal intellectual property rights hinder the free flow of scientific knowledge?: An empirical test of the anti-commons hypothesis
 +
*...
 +
*...
 +
*...
  
*Current directory containing candidate papers
+
Do the following:
*Processed BibTeX master file
+
*Create a new text file in the dropbox. Note where the file is saved!
 +
*Find each paper on Google Scholar and put the BibTeX entry into the text file
 +
*Add the fields from the sample entry above and file them out as follows:
 +
**Disciple: publication type
 +
**Research_type: 'Types' (see above)
 +
**Industry: industries (see above)
 +
**thicket_def_extract: The text from the paper that you will use to decide the thicket definition
 +
**thicket_def: one or more of DHCI, Saturated, Overlapping, Gaming, etc.
 +
**tags: hash tags for modern terms and topic key words, e.g., #Probabilistic, #PatentPool, etc.
 +
**Filename: Only download the file if we don't have it in '''Candidate Papers'''. Put the file name in this tag.
  
 
==Maybe Important Files==
 
==Maybe Important Files==

Latest revision as of 15:43, 2 September 2020

Academic Paper
Title Untangling the Economics of Patent Thickets
Author Ed Egan, David Teece
RAs Lauren Bass
Status Working paper
© edegan.com, 2016


Latest Progress

The PDF-to-Text converter, the key terms finder, and phrase extraction scripts all work well and have been tested. The Google Scholar Web Crawler is being worked on by Christy Warden, current status unknown.

192 text versions of papers have been assembled in "Candidate Papers by LB." These were used as a test set for developing an analysis protocol for the entire paper. An analysis protocol has been developed that involves taking the counts of each term by topic per paper to assign each paper 1-4 topics. The process also counts the number of key terms in each paper for relevance, the number of modern terms, the year of the paper, the number of cost terms, the area of the paper, and the mentions of particular authors. All of this analysis is then dumped into artifacts. The analysis protocol is currently hardcoded. Once a set list of terms has been established, a soft-coded version should be able to be developed (in progress).

Larger test set desired to be analyzed and soft-coded on.

The Paper

The latest version is:

\coauthoredprojects\Egan and Teece\McN-PatentThicket-Egan-092215.pdf

This file is posted at http://www.bakerinstitute.org/research/untangling-patent-thicket-literature

Codification

See the Patent Thicket Strategic Planning page for future steps.

See the PTLR Codification page for details

See the PTLR Process page for details

Test Run

Example BibTeX:

@article{BibTeXKeyGoesHere,
  title={},
  author={},
  journal={},
  year={},
  abstract={},
  discipline={},
  research_type={},
  industry={},
  thicket_stance={},
  thicket_stance_extract={},
  thicket_def={},
  thicket_def_extract={},  
  tags={},
  filename={}
}

Files

Pick 5 files at random from:

Dropbox\coauthoredprojects\Egan and Teece\CandidatePapers

List them here:

  • Cockburn MacGarvie (2006) - Entry And Patenting In The Software Industry.pdf
  • Murray Stern (2007) - Do formal intellectual property rights hinder the free flow of scientific knowledge?: An empirical test of the anti-commons hypothesis
  • ...
  • ...
  • ...

Do the following:

  • Create a new text file in the dropbox. Note where the file is saved!
  • Find each paper on Google Scholar and put the BibTeX entry into the text file
  • Add the fields from the sample entry above and file them out as follows:
    • Disciple: publication type
    • Research_type: 'Types' (see above)
    • Industry: industries (see above)
    • thicket_def_extract: The text from the paper that you will use to decide the thicket definition
    • thicket_def: one or more of DHCI, Saturated, Overlapping, Gaming, etc.
    • tags: hash tags for modern terms and topic key words, e.g., #Probabilistic, #PatentPool, etc.
    • Filename: Only download the file if we don't have it in Candidate Papers. Put the file name in this tag.

Maybe Important Files

  • PTLRCore.bib - Might contain core BibTeX entries
  • PTLRDefinitions.pdf - useful to contrast conflicting definitions
  • PTLRBibliography.bib - Long list of citations
  • PTLR-HoldupQuotes.txt - List of complete quotes with citations
  • PTLR-PolicyRpts.txt - Short list of policy reports ranging from 2002 - 2013
  • PTLRUp.bib - Another bibliography, additions or subtractions unknown
  • Files in C:\Users\Ed\Dropbox\coauthoredprojects\Egan and Teece\ThicketWiki
    • Particularly: CoreWithNewCoreForSorting-rev.txt

.PL files

  • BibTucker.pl - location PTLRv2 & 3 folders
  • PTLRBibTeXReprocessing.pl - location PTLRv2 & 3 folders
  • temp.pl - location PTLRv2 folder
  • BibTuckerV20.pl - PTLRv3 folder
  • PLTROldTables.txt