scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Patent pool formation and scope of patents

01 Oct 2011-Economic Inquiry (Blackwell Publishing Ltd)-Vol. 49, Iss: 4, pp 1070-1082
TL;DR: In this article, the authors investigate whether patent pools can rectify the lack of incentives for a developer to invest in an application, and they show that patent pools are more likely to provide enough incentives for basic innovators.
Abstract: Basic innovations are often fundamental to the development of applications that may be developed by other innovators. In this setting, we investigate whether patent pools can rectify the lack of incentives for a developer to invest in an application. Following Green and Scotchmer (1995), we also wonder whether broad basic patents are necessary in order to provide enough incentives for basic innovators. We show that patent pools are more likely

Summary (3 min read)

1 Introduction

  • In some areas of research (e.g., biotechnology), follow-up innovations may be built on several basic innovations, and might not be developed without them.
  • The authors show that patent pools are more likely to be formed either with patents of very di¤erent breadth or, on the contrary, with similarly broad patents.
  • If there is competition at the level of development, Scotchmer (1996) shows that patents on second generation products (when they infringe on the rst innovation) are not necessary to encourage their development, and the patentholder of the basic innovation collects a larger share of the pro t if second generation products are not patentable.
  • This seems to be consistent with one of their main ndings, that rms with broad patents are more likely to form a pool.

2 The Model

  • The associated (sunk) costs for discovering each basic innovation are c1 and c2.
  • In the extreme case of patents of in nite breadth, the application always infringes upon the patented basic innovations.
  • Qualitatively, it does not change their ndings, as explained in section 4.
  • Insert tree here First, both basic innovators decide simultaneously to form a patent pool or not.

3.1 Benchmark case: one basic innovation

  • Only rm A can develop the application.
  • Nevertheless, ex ante, the patentholder needs to have enough incentive to invest in the basic innovation.
  • For relatively low values of patent breadth (b1 2 [c1=K1v; k(v ca)=K1v]), the subgame perfect Nash equilibrium is such that the patentholder o¤ers an ex ante agreement that rm A refuses, but still invests in the application.
  • Indeed, if the value of the application v depends on the duration of the patent, and if the authors assume that the longer the patent, the higher the value of the application, long patents do not need to be broad.

3.2 Two Basic Innovations: Infringement or No Infringement

  • The authors now consider two basic patents held by two innovators.
  • However, if there is infringement, ex post e¢ ciency must be achieved and, therefore, the developer will invest only if (1 ki)((1 Kj)v ca) > 0; i.e., less often than in the case with only one patentholder.
  • If assumption (2) is relaxed, there exist two symmetric Nash equilibria.
  • Proposition 2 Under assumptions (1)-(5), if the application infringes on both patents, there exists a Nash equilibrium in which the two patentholders decide to pool.
  • The developer accepts the ex ante agreement.

3.3 Breadth of Patents

  • The authors now consider that the probability of infringing is strictly positive.
  • If only one patentholder proposes an ex ante agreement, the developer will get (1 ki)((1 bjKj)v ca) and, thus, the investment will be undertaken only if (1 bjKj)v ca >.
  • Absent a patent pool, patentholders with strong monopoly rights would not be able to enjoy those rights, as no application would be developed.
  • For intermediate values of breadth, the basic innovator who proposes an ex ante agreement does not have enough incentive to invest.
  • The broader the patent, the higher the probability of free-riding.

4 Extensions

  • The authors now explore two extensions of the model.
  • In the rst one, the authors consider that in case of patent pool failure, each patentholder can independently o¤er an ex ante agreement.
  • Second, the authors consider a sequential game instead of a simultaneous one, in which the rst patentholder rst o¤ers a pool, and the second patentholder decides whether to join the pool or not.

4.1 Patent Pool versus Individual Ex Ante Agreements

  • The timing of the game is now the following: First, the two patentholders decide simultaneously whether to pool their patents or not.
  • In case of patent pool failure, each patentholder decides simultaneously to o¤er separate ex ante agreements to the developer.
  • To solve this new game, the authors rst solve the ex ante agreement subgame before solving the patent pool subgame.
  • If the authors make the realistic assumption that kpi > k a i , i.e., that the share of payo¤ that each patentholder will receive from being part of a pool is higher than the share of payo¤ each will get from o¤ering independent ex ante agreements, the results are identical to those obtained in the previous section.
  • Therefore, in this simple model, because the authors do not explicitly determine the optimal share of payo¤ of each of the patentholders, o¤ering patent licenses in a pool dominates o¤ering separate licenses.

4.2 Sequential Formation of a Pool

  • The authors may also be concerned with the fact that pools in general are formed sequentially rather than simultaneously.
  • One rm may decide to o¤er a pool with one patent, and another rm may decide to join the pool after its creation.
  • In their setting, this change in the sequence of decisions does not a¤ect their ndings.
  • Indeed, pools will emerge with the same characteristics as those the authors de ned in proposition.
  • This is again due to the fact that the shares of the payo¤ are exogenous.

5 Conclusion

  • Patents play an important role in many industries (e.g., biotechnology), and the existence of too many property rights on basic innovations impacts follow-up applications.
  • All of the variables are exogenous, and the only decisions are on whether to invest and whether to pool.
  • The authors do not consider any strategic choice of breadth, but, rather, they investigate all of the possible pools that will emerge depending on the di¤erent breadths.
  • Thus, one may wonder if the emergence of patent pools that has been observed during recent years could be the result of a patent policy that allows for broad patents.
  • From an antitrust standpoint, such agreements are less problematic since they do not result in a strong barrier to entry, but still foster and restore incentives to develop applications.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

Patent Pool Formation and Scope of Patents
Corinne Langinier
September 2007
Abstract
Basic innovations are often fundamental to the development of applications that may
be developed by other innovators. In this setting, we investigate whether patent pools can
rectify the lack of incentives for a developer to invest in an application. Following Green
and Scotch mer (1995), we also wonder whether broad basic patents are necessary in order to
provide enough incentives for basic innovators. We show that patent po ols are more likely
to b e formed with patents of very di¤erent breadths, or patents of similarly wide breadths.
Further, although patent pools rectify the problem of developer incentives, they may reduce
the incentive for doing basic research.
Keywords: Patent pool, innovation, breadth
JEL classi…cation: K11, L4, O31
Iowa State University, Ames. Tel: (515) 294-5830; Fax: (515) 294-0221 langinier@econ .iastate.edu
I am indebted to comments from Philippe M arco ul, Jorge Andres Ferrando Yanez and other participants at the
IIOC 2004 in Chic ago.
1

1 Introduction
In some areas of research (e.g., biotechnology), follow-up innovations may be bu ilt on several
basic innovations, and might not be developed without them. Inventions such as methods to
isolate and locate gene sequences have no value by themselves, but they permit the development
of subsequent valuable applications. Furthermore, basic innovations and applications are usually
developed by di¤erent companies.
1
Two important incentive problems emerge from sequential inn ovations. First, basic innova-
tors must be given enough incentive to promote their innovations. The literature on sequential
innovation has mainly focused on how to protect the rst generation of innovators against future
innovators. In other words, it is concerned with the transfer of pro…t from second generation
innovators to the initial innovators in order to promote basic innovations (Scotchmer, 1996;
Green and Scotchmer, 1995; Chang, 1995). First gen eration innovators should be given broad
protection when second generation innovations can only be obtained by an outsider (Green and
Scotchmer, 1995).
2
Second, follow-up innovations may not be brought about when developers decide not to
pursue research if it is built on several basic innovations (Merges and Nelson, 1994). Indeed,
if basic patents are too broad, follow-up innovators may have to pay too many fees to be able
to develop applications. This phenomenon, referred to by Heller and Eisenberg (1998) as the
tragedy of anticommon,” depicts a s ituation in which there are too many door keepers, an d
to build on previous innovations an innovator needs the permission of too many patentholders
(Shapiro, 2001). This second problem is specially acute in biotechnology. For instance, a survey
of laboratory physicians shows that because of patents one fourth have abandoned a clinical test
that they have developed, and almost one half report that they have not developed a test for
fear that they would be sued.
3
In biotechnology, and more gene rally in elds where innovation is cumulative and pyramidal,
1
For instance, a public lab oratory o r a research department at a university can develop basic innovati ons, while
letting the private sector develops a pplications.
2
In the case of a long sequence of innovation, the optimal scope of patents is provided to stimulate R&D
investment (O’Donoghue, Scotchmer and Thisse, 1998).
3
This survey has been conducted by Jon Merz and Mildred Cho, bioethicists at the University of Pennsylvania,
and has been reported by Thompson, Washington Monthly, April 2001.
2

one of the solutions is to license patents in a patent pool. A patent pool is an agreement between
two or more patentholders to license one or more of their patents to one another or to third
parties (Merges, 1999).
In this paper, we examine how a patent pool can rectify the lack of developer incentives
to invest in an application when basic innovators cannot develop follow-up applications. This
situation is relatively common, as basic innovators may be universities or public research labora-
tories that do not have the ability to develop applications. We investigate whether broad basic
patents are necessary in order to provide enough incentive to basic innovators, as is the case in
the Green and Sc otchmer (1995) model. Further, we wonder wh at kind of pool will emerge. In
other words, are patents more likely to be broad or narrow in a patent pool? We consider a
model in which there are two patentholders of basic innovations that are mainly research tools,
with null market values, but that permit a third rm to develop an application. In this setting,
we show that patent pools are more likely to be formed either with patents of very di¤erent
breadth or, on the contrary, with similarly broad patents.
In 2001, several companies (GE Healthcare, f ormerly Amersham Biosciences, Biolmage A/S
and Invitrogen IP Holdings, formerly Aurora Biosciences Corporation) and Colombia University
agreed to pool se veral of their patents on green uorescent protein (GFP). Th is is a uorescent
reporter molecule used in drug discovery to create a detailed picture of how potential drugs ect
the function of protein. The patent pool contains several U.S. patents, as well as European and
Japanese patents. All users of GFP are required to obtain a license to use the technology prior
to starting research work”(GE Healthcare conditions for licensing).
4
Therefore, one can consider
a patent pool as an ex ante agreement ered to any potential developers. The application areas
include research in cell biology and pharmaceu tical screening. Using Lerner’s proxy of the scope
of patents, i.e., the number of the rst four digits of the International Patent Classi…cation
(IPC) in each patent, we nd that the patents included in the pool tend to be broad. Indeed,
according to Lerner’s analysis (1994), the number of four-digit IPC classes into which a patent
is classi…ed is, on average, 1.68 for biotechnology patents. Among the eleven patents that are
in the pool, if we put together those that are continuations of others (two patents, each with
two continuations), most have more than two four-digit IPC classes. In fact, only one of them
has one four-digit IPC class. This seems to be consistent with our nding that pools are more
4
See the web site http://www.gehealthcare.com.
3

likely to be formed with broad patents (see Table 1 that summarizes the information about this
existing patent pool).
Our analysis is related to the two streams of literature mentioned above: the literature on
sequential innovation, especially the paper by Green and Scotchmer (1995), and the more recent
literature on patent pools.
In a model with one patentholder of a basic innovation and one potential innovator to develop
the application, Green and Scotchmer (1995) show th at an ex ante agreement should be allowed
in order to insure investment in the second innovation. When the re is no uncertainty about
the value of the follow-up innovation, the best policy consists in giving a broad patent to the
rst innovator, and the ex ante licensing improves social welfare whatever the patent breadth.
5
However, if the d evelopers’costs are private knowledge, patentholders do not necessarily er
ex ante licensing (Bessen, 2004). We built upon the model of Green and Scotchmer (1995) to
investigate patent pools. We explicitly introduce patent breadth, and we extend their analysis
to two basic patents that can be pooled in an ex ante agreement in a model where there is
common knowledge about costs. The main objective of the Green and Scotchmer (1995) model
is to study patent policy and to determine what the optimal patent breadth should be. In our
setting, we show how given patent breadths impact the choice of whether or not to form a patent
pool. Therefore, our focus is on strategic choices by rms to f orm a patent pool as a function of
patent breadth.
Patent po ols are viewed as a possible solution for reducing additional transaction costs
incurred in navigating the patent thicket (Shapiro, 2001) or the anticommon problem (Heller
and Eisenberg, 1998; Buchanan and Yoon, 2000). They can also enhance ciency by eliminating
the complement problem (Shapiro, 2001), and may bene…t society (Gilbert, 2004; Lerner and
Tirole, 2004; Gilbert and Katz, 2006). Following Cournot analysis, Sh apiro (2001) shows that a
package of licenses for basic innovations neither harms consumers nor the rms themselves, as
long as the royalty rates are low.
5
If there is competition at t he level of development, Scotchmer (1996) shows that patents on second generation
products (when they infringe on the rst innovation) are not necessary to encourage th eir development, and the
patentholder of the basic innovation collects a larger share of the prot if second generation pr oducts are not
patentable. On the other hand, if competition can take place at both research and development stages , and if
both stage s can be done by the same rm, Denicolo (2000) s hows that weak forward prot ection can be preferable.
4

The major bene…ts of patent p ools are that they eliminate staking licensing, reduce licensing
transaction costs, reduce patent litigation and allow for the exchange of information. However,
a patent pool can have anti-competitive ects. Antitrust authorities have been consistently
more suspicious of a po ol of substitutable p atents than of a pool of complementary patents. If
patents are perfect complements, a pool eliminates the double marginalization (Shapiro, 2001).
If they are perfect sub stitutes, a pool eliminates competitors. However, patents are rarely perfect
complements or perfect substitutes. In this context, Lerner and Tirole (2004) study a model of
pool formation and pricing, and show that a pool can increase or decrease the price, depending
on the internal or external competition. Further, large pools are more likely to allow individual
licensing, to centralize control of litigation and to license to third parties (Lerner, Strojwas and
Tirole, 2003). In our model, we consider that a patent pool is an ex ante agreement where
patents are complements, as both innovations are needed to develop the application. In this
setting, we wonder what breadths of patents are more likely to be pooled together.
In most of the literature on patent pools, it is assumed that rms do participate in the pool
(e.g., Lerner and Tirole, 2004). However, joining a pool is a voluntary decision and, therefore,
rms may decide not to join a patent pool. In the setting of a consortium standard, Aoki and
Nagoaka (2004) analyze the decision of a rm not to join a patent pool, or the reasons for
the split of a patent pool. They show that because of free-riding and bargaining failures (due
to heterogenous rms), rms do not always want to be part of a patent pool. In our setting,
our ndings are similar: when rms are heterogenous (in terms of patent breadths), a patent
pool can fail. However, in the extreme case, when rms have patents of very di¤erent breadth,
we nd that a pool will be formed. The apparent contradiction in our ndings is due to the
di¤erent modelling framework. Indeed, in their model, implicitly ex post ciency is always
achieved, whereas we do account for situations in which the application will not be developed.
From an empirical standpoint, Layne-Farrar and Lerner (2006) provide an analysis to test the
determinants of joining a pool. One of their ndings is that rms with relatively symmetric
contributions to a new technology are more likely to form a patent pool. This seems to be
consistent with one of our main ndings, that rms with broad patents are more likely to form
a pool. Our paper is a contribution to this new stream of literature on patent pools that is
related to the voluntary aspect of joining a patent pool. In a setting where the scope of patents
is acc ounted for, we study the incentives to join a pool and analyze the formation of patent
5

Citations
More filters
Journal ArticleDOI
TL;DR: In this paper, the authors investigated the patent holders' incentive to form a patent pool, the patent pool's licensing behavior, and the anticompetitive effect of such a pool.
Abstract: Patent pools are established by the patent holders in order to promote R & D and technological standards, etc. This paper investigates the patent holders’ incentive to form a patent pool, the patent pool’s licensing behavior, and the anticompetitive effect of a patent pool. Our model is characterized by the following two features. First, we consider the different two types of patent: basic and optional. Second, we consider a patent pool that offers two types of a package license: single and multiple. Our results yield some implications for a patent pool that is characterized by the complementarity between basic and optional patents.

1 citations

Journal Article
TL;DR: This paper investigates the anticompetitive effects of a patent pool that offers a package license to users and develops a model that can analyze the multiple package licenses offered by a patent Pool and discusses multiple package license from an antitrust perspective.
Abstract: Patent pools are organizations where patent holders concentrate their own patents and offer licenses to each other and third parties Most of the literature on patent pools has analyzed the single package license, which includes all the patents in the pool (Lerner and Tirole 2004; Shapiro 2001) However, to date there has been no study of multiple package licenses, which are packaged within subsets of all the patents in the pool This paper develops a model that can analyze the multiple package licenses offered by a patent pool and discusses multiple package licenses from an antitrust perspective This paper investigates the anticompetitive effects of a patent pool that offers a package license to users Our analysis is characterized by two types of package licenses: the single package license and the multiple package license The single package license is inclusive of all patents in the patent pool If the single package license is offered, users can use all of the patents in the patent pool to commercialize new innovations The multiple package license includes a subset of the patents in the patent pool If the multiple package license is offered, users can select a license that only includes the patents they require from the patent pool A patent pool refers to organizations where patent holders concentrate their own patents for commercializing new innovations or for setting standards, and offer a package license that is inclusive of many of the patents in the pool

1 citations

Journal ArticleDOI
Ozgur Aydogmus1
TL;DR: In particular, the authors analyzes patent pool dynamics using nonlinear Polya urns and shows that increasing the number of firms contributing to a patent pool of complementary patents or increasing the rate of having infringing patents favor the contributing firms' coexistence chance.
01 Nov 2013
TL;DR: In this article, the formation characteristics of patent pools under technical standards are studied. But the authors focus on the information technology industry, union of interests among companies with shared advantages, and formation of patent pool that are actively promoted by standard-setting organizations and third-party administrators.
Abstract: How do enterprises build an appropriate patent pool when technical standards are regarded as prerequisites and bases to establish them? This study addresses the issue by examining the technical standards and relative process implications of de facto standards and statutory standards. Based on statistics and research on patent pools with 34 technical standards from 1993 to 2012, this paper emphasizes on three formation characteristics of patent pools under technical standards: focus on the information technology industry, union of interests among companies with shared advantages, and formation of patent pools that are actively promoted by standard-setting organizations and third-party administrators. The formation process of the DVD and MPEG patent pools largely support this theory.
References
More filters
Journal ArticleDOI
01 May 1998-Science
TL;DR: Privatization of biomedical research must be more carefully deployed to sustain both upstream research and downstream product development, because more intellectual property rights may lead paradoxically to fewer useful products for improving human health.
Abstract: The “tragedy of the commons” metaphor helps explain why people overuse shared resources. However, the recent proliferation of intellectual property rights in biomedical research suggests a different tragedy, an “anticommons” in which people underuse scarce resources because too many owners can block each other. Privatization of biomedical research must be more carefully deployed to sustain both upstream research and downstream product development. Otherwise, more intellectual property rights may lead paradoxically to fewer useful products for improving human health.

2,371 citations

Journal ArticleDOI
TL;DR: In this paper, the impact of patent scope on firm value was examined using a sample of 535 financing rounds at 173 privately held venture-backed biotechnology firms, and it was shown that a one standard deviation increase in average patent scope is associated with a 21% increase in the firm's value.
Abstract: This article examines the impact of patent scope on firm value. I develop a proxy for patent scope based on the International Patent Classification scheme. Using a sample of 535 financing rounds at 173 privately held venture-backed biotechnology firms. I show that the breadth of patent protection significantly affects valuations. A one standard deviation increase in average patent scope is associated with a 21% increase in the firm's value. Broad patents are more valuable when substitutes in the same product class are plentiful, a finding consistent with theoretical suggestions.

1,043 citations

Journal ArticleDOI
TL;DR: The patent thicket is especially thorny when combined with the risk of hold-up, namely the danger that new products will inadvertently infringe on patents issued after these products were designed as mentioned in this paper.
Abstract: In several key industries, including semiconductors, biotechnology, computer software, and the Internet, our patent system is creating a patent thicket: an overlapping set of patent rights requiring that those seeking to commercialize new technology obtain licenses from multiple patentees. The patent thicket is especially thorny when combined with the risk of hold-up, namely the danger that new products will inadvertently infringe on patents issued after these products were designed. The need to navigate the patent thicket and hold-up is especially pronounced in industries such as telecommunications and computing in which formal standard-setting is a core part of bringing new technologies to market. Cross-licenses and patent pools are two natural and effective methods used by market participants to cut through the patent thicket, but each involves some transaction costs. Antitrust law and enforcement, with its historical hostility to cooperation among horizontal rivals, can easily add to these transaction costs. Yet a few relatively simple principles, such as the desirability package licensing for complementary patents but not for substitute patents, can go a long way towards insuring that antitrust will help solve the problems caused by the patent thicket and by hold-up rather than exacerbating them.

771 citations

Journal ArticleDOI
TL;DR: In this paper, the authors investigate the policy that is most effective at ensuring the first innovator earns a large share of profit from the second-generation products it facilitates, and propose cooperative agreements between initial innovators and later innovators.
Abstract: In markets with sequential innovation, inventors of derivative improvements might undermine the profit of initial innovators through competition. Profit erosion can be mitigated by broadening the first innovator's patent protection and/or by permitting cooperative agreements between initial innovators and later innovators. We investigate the policy that is most effective at ensuring the first innovator earns a large share of profit from the second-generation products it facilitates. In general, not all the profit can be transferred to the first innovator, and therefore patents should last longer when a sequence of innovations is undertaken by different firms rather than being concentrated in one firm.

724 citations

Journal ArticleDOI
TL;DR: In this article, the trade-off between a patent's length and its scope of coverage is explored, and it is shown under what conditions infinitely lived but very narrowly focused patents are the socially efficient way to reward innovation.
Abstract: I explore the trade-off between a patent's length (that is, its lifetime) and its width (that is, its scope of coverage). A wider patent generally reduces the distortion of consumers' choices between the patented brand of the product and unpatented, lower-priced varieties sold by competitors but also permits higher prices, which increase (relative to profits) the deadweight losses from consumers switching consumption out of the product class. I show under what conditions infinitely lived but very narrowly focused patents are the socially efficient way to reward innovation and under what conditions very short-lived but very broad patents are optimal.

691 citations

Frequently Asked Questions (1)
Q1. What are the contributions mentioned in the paper "Patent pool formation and scope of patents" ?

In this setting, the authors investigate whether patent pools can rectify the lack of incentives for a developer to invest in an application. Following Green and Scotchmer ( 1995 ), the authors also wonder whether broad basic patents are necessary in order to provide enough incentives for basic innovators. The authors show that patent pools are more likely to be formed with patents of very di¤erent breadths, or patents of similarly wide breadths. Further, although patent pools rectify the problem of developer incentives, they may reduce the incentive for doing basic research.