# Is Patent Reform via Private Ordering Anticompetitive? An Analysis of Open Patent Agreements

**Authors:** Matthew W. Callahan, Jason M. Schultz
**Citation:** Is Patent Reform via Private Ordering Anticompetitive? An Analysis of Open Patent Agreements, in *Patent Pledges: Global Perspectives on Patent Law's Private Ordering Frontier* 151 (Jorge L. Contreras and Meredith Jacob, eds., Edward Elgar, 2017)
**Source:** https://www.elgaronline.com/edcollbook/edcoll/9781785362484/9781785362484.xml

*p. 2*
with every additional party. Yet whenever networks of actors agree on specific obligations or actions, especially ones that hold interests in relevant intellectual property, such an agreement also implicates competition policy -and liability. This chapter contemplates this topic in the context of United States federal antitrust law. Specifically, it looks at how antitrust regulators and doctrines will likely react to OPAs, both in their nascent forms and over time if they are successful at attracting large networks of users. OPAs, as we define them, involve efforts to license patents broadly to all-comers for royalty-free use, usually on certain conditions that aspire to promote widespread innovation and reduce patent litigation. 6 No court or regulator has directly confronted the issue of whether these OPAs would violate antitrust laws. By looking at the related court decisions and regulatory guidance on patent pools and open source copyright licenses, Section II of this chapter attempts to build a framework for evaluating patent pledges generally. Section III of this chapter applies this framework to OPAs, concluding that OPAs should survive most antitrust challenges under existing U.S. law.

## A. Patent Pools and OPAs

*p. 2*
Collective action via patent agreements is nothing new. Patent pools have been around in the United States since at least the mid-1800s, starting 6 See Eric Adler, The Basics of Open Patent Licensing, TECHCRUNCH (Apr. 23, 2015). While exact statistics are not available, the OPA governing the largest number of patent grants is probably that contained in software licensed under free and open source software licenses. Someone who contributes code under the terms of the GNU General Public License (GPL) pledges not to enforce any patent rights covering that code against subsequent licensees (implicitly in older versions of the license, and explicitly in later ones). See GNU General Public License version 2.0 § 6 (1991), https://www.gnu.org/licenses/gpl-2.0.html#SEC1 (containing the implicit patent grant); BRADLEY M. KUHN & ANTHONY K. SEBRO, JR., COPYLEFT AND THE GNU GENERAL PUBLIC LICENSE: A TUTORIAL 28-29 (2014) (explaining the implicit patent grant in GPLv2); GNU General Public License version 3.0 § 11 (2007), https://www.gnu.org/licenses/gpl-3.0.en.html (containing the explicit patent grant). However, because the GPL is primarily a copyright license with so-called "patent peace" provisions in it, it is beyond the scope of this chapter to consider.

*p. 3*
with the Sewing Machine Combination (1856-77). 7 More recent examples include patent pools for video formats, mobile network technology, and even diagnostic genetic testing. 8 Pools allow multiple parties to combine their patents for a variety of pro-innovation purposes, such as an efficient means to resolve ongoing patent disputes and to lower transaction costs for cross-licensing blocking patents among repeat-player competitors. 9 Applicable patents are usually defined by their economic and technological roles, that is, as substitutes or complements within a particular technological domain. However, as with many collective action arrangements, competition concerns inevitably emerge. Still, regulators and courts have allowed such pools to exist when, "in a case involving blocking patents, such an arrangement is the only reasonable method for making the invention available to the public[.]" 10 Pools may also encourage firms to enter the innovation race on a particular technology so they can be the first to patent key aspects and maximize their individual returns from the pool's profits. 11 On the other hand, this incentive to "race" may be misplaced or even socially wasteful if it "encourages duplicative research and strategic patenting." 12 Moreover, once a pool is formed, it can lessen the economic and cultural incentives among participants to invest in research and development of new technologies. After all, if everyone in the pool gets a share of the pie, why spend the money to develop new recipes and bake new pies? 13 Conversely, for competitors outside the pool, there can be greater incentives to "invent around" pool-covered technologies in order to avoid higher licensing fees which make it more difficult to offer technologies covered by the pool at a price competitive with pool members. Id. (noting that incentives to free-ride are particularly strong for pools that include "grantback provisions," which require members to offer all new patents to the pool, and innovative members may abandon the pool to protect their patents) (citing Reiko Aoki & Sadao Kagaoka, The Consortium Standard and Patent Pools, 55 ECON. REV. 345 (2004).

*p. 4*
Open patent agreements exhibit many of the same characteristics as patent pools, but not all. They are often focused on a particular technology or technological standard with the same intent to reduce litigation costs/risks and to lower cross-licensing transaction costs. However, unlike traditional patent pools, they typically offer licenses to patents on the basis of pledges or conditions related to "openness"generally understood as encouraging inclusiveness in the development of technologies instead of intellectual property's traditional exclusive emphasis. For example, the Open Invention Network (OIN) offers royalty-free licenses to its entire patent portfolio on the Linux operating system as long as each licensee agrees to grant a similar license for its patents on Linux to every other licensee in the network. As such, it holds itself out as "a shared defensive patent pool with the mission to protect Linux" 14 that was created to "ensure a level playing field for Linux, safeguarding developers, distributors and users from organizations that would leverage intellectual property to hinder its growth and innovation." 15 Most importantly, it is free to join the network as long as one commits to the cross-licensing model.

*p. 4*
The License On Transfer (LOT) network also focuses on safeguarding its members from patent litigation through an "open" approach to licensing. To join LOT, members agree that, if they ever transfer ownership of any of their patents in the future, every member of the network at that time will receive a royalty-free license to every transferred patent. In theory, this discourages so-called "patent trolls" from acquiring any LOT patents, as the act of acquisition will immunize network members from any future threats from those patents. The cost of joining LOT varies, but the costs are largely administrative and proportional to the size and revenue of the joining entity. 16 The Defensive Patent License (DPL), co-authored by one of the authors of this chapter, also focuses on leveraging licensing rights in favor of openness and litigation reduction. 17 technological focus. DPL licensors agree to offer royalty-free licenses to their entire patent portfolios, and licensees, in turn, agree never to assert any of their patents offensively against any member of the DPL network. If a licensee breaches, each licensor has the option of terminating the license. If a particular licensor wishes to cease offering DPLs to new network members, they must give the network six months' notice. Upon leaving, the former Licensor's DPLs can be converted into fair, reasonable, and non-discriminatory (FRAND) royalty-bearing licenses.

*p. 5*
In looking at these OPAs, certain commonalities emerge related to competition and innovation analysis. First, they place no restrictions on additional licensing activity either inside or outside of their networks; any network member can strike customized deals with any other network or non-network member. If two members of the OIN network wish to enter into a joint distribution deal involving certain patented products, the OIN license would not preclude a separate side license as part of that transaction as long as it did not breach the OIN prohibition on infringement actions against developers, distributors, or users of the Linux system. Second, OPAs do not place any restrictions on licensing or litigation behavior outside of the network. Third, because the cost of joining most OPAs is free or, at most, administrative, the barriers to any new entrant are extremely low.

## B. Antitrust Law and Legal Frameworks

*p. 5*
Antitrust law is, famously, divided into business practices which are per se illegal and those which are evaluated under the "rule of reason." 18 Courts have found per se antitrust violations in cases where, after long experience, the practice restrains trade in almost all of its occurrences. 19 Without such a history, the Supreme Court has expressed reluctance to find a per se antitrust violation "where the economic impact of certain practices is not immediately obvious." 20 Business practices that are not per se violations are evaluated instead under the rule of reason. When courts look at an activity under the rule of reason, they take into account "a wide variety of factors, including specific information about the relevant business, its condition before and after the practice was imposed, and the practice's history, nature, and effect." 21 The rule of reason thus provides a more forgiving, case-by-case analysis than the per se violations.

*p. 6*
Given that OPAs are a relatively new phenomena and one where the economic impact is far from certain, OPAs will likely be evaluated under the rule of reason. This is reinforced by the fact that the United States Department of Justice (DOJ) and the Federal Trade Commission (FTC) have both stated that they generally evaluate patent pools under the rule of reason, 22 and the United States Court of Appeals for the Federal Circuit has likewise stated that the rule of reason is appropriate when licensing agreements are related to a larger collaborative effort. 23 Moreover, in evaluating an antitrust challenge to the open source licensing of the Linux operating system, the Seventh Circuit applied the rule of reason, finding that the General Public License (GPL) used was reasonable under that test. 24 Under the rule of reason, a court or regulator must still confront the question of what factors to consider. The rule of reason is difficult to apply in practice and the Supreme Court has provided little guidance on the issue. 25 administrative agencies charged with enforcing antitrust law, most notably the DOJ and the FTC. 26 As a result, guidance from the DOJ and the FTC provides the best starting point for which factors to evaluate under the rule of reason in determining whether OPAs violate antitrust law. In particular, while OPAs are not the same as patent pools, they do involve the collective action of multiple patent holders and share many of the attributes of OPAs that would interest regulators. Because of this close relationship, and because the antitrust considerations about patent pools are relatively well developed, we look to the DOJ and FTC guidance on patent pools to provide some basis for OPA antitrust analysis.

*p. 7*
The DOJ/FTC guidance identifies two main concerns regarding patent pools: (1) that "horizontal coordination among the pool's licensors could lead to a reduction in price competition among downstream products"; and (2) that "combining patent rights in a pool could discourage R&D, new product development, and cost-reducing process innovations." 27 In evaluating specific patent pools, the agencies have looked at factors such as whether, in these pools, "licensors would retain the right to license their patents individually, the scope of grant-back clauses would be limited, the license agreement would be available to all interested licensees, and the pool would provide a clear understanding of the contents of the license." 28 The main judicial opinion to evaluate open source copyright licenses under the rule of reason also considered such concerns, largely finding them absent in the open source software context. In an opinion by Judge Frank Easterbrook, the Seventh Circuit held, " [t]he GPL and open-source software have nothing to fear from the antitrust laws" 29 and that "the GPL does not restrain trade. It is a cooperative agreement that facilitates production of new derivative works, and agreements that yield new products that would not arise through unilateral action are lawful." T]he GPL in no way forecloses other operating systems from entering the market. Instead, it merely acts as a means by which certain software may be copied, modified and redistributed without violating the software's copyright protection. As such, the German courts have agreed. 31 While some scholars have asserted viral components of the GPL and other viral copyleft licenses have been called anticompetitive, even the scholars making this claim doubt that current antitrust law would find them illegal without legislative reform. 32 With these frameworks in mind, we now turn to the question of OPAs, competition, innovation, and antitrust liability.

## A. Horizontal Coordination and Price Competition 1. Horizontal coordination

*p. 8*
The DOJ and FTC guidance expresses concern that "horizontal coordination among the [patent] pool's licensors could lead to a reduction in price competition among downstream products[.]" 33 The main threat of "horizontal coordination" is that it requires patent holders, many of whom may be competitors in the same market, to come to agreements about pricing, at least at the early stages of development and distribution.

*p. 8*
Preventing horizontal competitors from colluding to extract rent at the expense of consumers is one of the core concerns of antitrust law. 34 In looking at horizontal coordination, it is instructive to contrast OPAs with another kind of patent pledge: Defensive Patent Aggregators (DPAs). DPAs are non-practicing entities that license and acquire patents on behalf of member companies in order to reduce the future uncertainty of litigation and reduce the cost of any potential liability. For instance, DPA Allied Security Trust (AST) implements what is called a "catchand-release" defensive acquisition approach. It requires that companies GPL encourages, rather than discourages, free competition and the distribution of computer operating systems, the benefits of which directly pass to consumers. … [Plaintiff] does not adequately set forth an injury to competition as a whole").

*p. 8*
LG Frankfurt, Sept. 6, 2006, No. 2-6 O224/06 (13) (Ger.) (holding GPL was enforceable, even though it set the price for alterations of software at zero, not an antitrust violation) ( pay it a subscription fee. 35 In exchange, it coordinates anonymous negotiations among its member companies about how much they are willing to contribute to the acquisition of certain patents and, based on these negotiations, it purchases a patent. 36 All parties to the bid are given licenses; then the patent is resold on the open market, and the profits are divided among the bidding companies. 37 RPX Corporation, another DPA, provides licenses to subscribing companies for as long as they are members. 38 RPX seeks to purchase either patents or licenses for its members, including patents that are involved in active litigation. 39 RPX is very large, claiming to represent 10% of patent transactions on the open market. 40 These two DPAs have faced antitrust litigation. Allied Security Trust was sued on claims that it conspired with its members to fix prices for a set of patents; however, the court did not reach the merits of the claims, instead dismissing the case due to lack of antitrust standing. 41 RPX Corporation was sued in federal court on similar claims that it was part of a "hub-and-spoke conspiracy" to monopsonize a patent market by convincing all RPX subscribers to agree not to pay above a certain price for patents owned by the plaintiff. 42 The case was later dismissed by the court after a jury found the defendants did not infringe on the plaintiff's patent. 43 clear that some risk exists in coordinating networks to agree on paying or bidding a particular price for a particular patent. 44 OPAs, however, lack any specific means or mechanism for horizontal coordination of purchases or bidding on patents. In fact, the standard language of all OPA agreements suggests no real benefit to coordination, as the terms of any OPA-related transaction will be identical regardless of the party, patent, or technology involved. Communication between parties, other than through the mechanism of the license, is neither required nor provided for. This does not mean it is impossible that two users of an OPA would conspire to fix prices horizontally, but only that being users of the OPA does not make it any more or less likely to happen. The users of the OPA will be just as guilty, and just as susceptible to proof of guilt, as two conspirators who are not part of the OPA. The kind of coordination that gives RPX, AST, or other actors the potential to set prices, either as purchasers or sellers, never occurs with OPAs.

## Price competition

*p. 10*
Even assuming horizontal coordination among OPA patent holders were possible, a subsequent decrease in price competition is also unlikely. 45 This is for two reasons. First, because barriers to joining OPA networks are extremely low or nonexistent, OPA network members cannot rely on OPAs to maintain monopoly pricing; if they did, outside competitors would simply enter the market and take whatever OPA patents were required to compete. Second, any attempt to reduce competition through "predatory pricing" of OPA patents would fail, as predatory pricing only occurs when there is the possibility of a subsequent rise in prices. 46 The Seventh Circuit, in an opinion by Frank Easterbrook dismissing a predatory pricing claim against IBM for distributing software under the GPL, stated "[w]hen [competitor] exit does not occur, or recoupment is improbable even if some producers give up the market, there is no antitrust problem." 47 Thus, the requirement of royalty-free licensing in 44 See also Samsung Elecs. Co. v. Panasonic Corp., No. C 10-03098 JSW, 2015 WL 10890655, at *5 (N.D. Cal. Sept. 30, 2015) ("Anticompetitive effects may arise from patent pooling arrangements that require payment for a pool of rights without a realistic opportunity as a practical matter to obtain individual licenses from individual owners as opposed to a single license from the pool"); Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co., 549 U.S. 312, 315 (2007) (holding that predatory pricing theory applies to predatory bidding).

*p. 11*
OPAs should not raise price competition problems since there is no strategy for recoupment in the form of future profits.

*p. 11*
If anything, parties breaching OPAs could open themselves up to antitrust liability. Prof. Jorge Contreras has suggested that patent pledgers inducing other parties to rely on a pledge and then abandon it may violate antitrust law under a "market reliance" theory. 48 The FTC in recent years has shown an interest in bringing suit against deceptive practices in the patent space. For instance, in 2013, the FTC accepted a consent decree over a proposed merger which involved a company that had promised certain standard-essential air conditioning patents on FRAND terms and allegedly breached that promise. 49

## B. Discouraging Innovation

*p. 11*
The other major concern of the DOJ and FTC in evaluating patent pools is that they "could discourage R&D, new product development, and cost-reducing process innovations." 50 This invites the consideration of the effects of the OPA licensing models on innovation generally.

*p. 11*
Open patent agreements operate very differently from traditional patent pools in that their members do not receive any direct revenue from the agreements but do benefit in the reduction of litigation uncertainty and expense as well as reduction in the payment of potential licensing fees. The lack of direct revenue could, in theory, discourage research and development (R&D), new product development, and other cost-reducing process innovations. However, as we discuss below, the benefits of OPAs directly encourage all three of these activities thus outweighing any concerns that might exist regarding these factors. All OPAs provide royalty-free licensing opportunities but only when all their other conditions are met. Rather than discouraging R&D among licensees, such opportunities encourage it. By removing the fear, uncertainty, and doubt of litigation among network members, each entity has the incentive to develop complementary and follow-on technologies to any patented product or service they have licensed. Moreover, because they know that any of their competitors will be able to enter the network and license any of their patents for free at any time, network members will have incentives to be constantly innovating ahead of their patent portfolio, so as to keep first-mover advantage on any new innovations. Investments in R&D will be further protected through the use of non-disclosure agreements (NDAs) on unpatented trade secrets and "know-how" techniques, especially those that involve cost-reducing processes. Moreover, OPA network members continue to have the concurrent opportunity to generate revenue through licenses outside the networks.

*p. 12*
1. OPAs encourage R&D through traditional licensing outside of their networks Firstly, OPAs are not designed to compete with or replace traditional revenue-generating licensing markets. Rather, they are meant as complements to those licenses and to facilitate competition at the products and services layer. For instance, the GNU/Linux operating system licensed under the GPL competes in the computer operating system market with closed source, proprietary operating systems produced by companies like Microsoft and Apple. 51 Developers of products and services often acquire licenses for each operating system in order to maximize adoption across systems. The balance between open and closed models has proven so beneficial for consumers that when a Microsoft-led consortium sought to acquire patents essential for the operation of the GNU/Linux operating system, the DOJ wrote a letter to the consortium insisting that the acquired patents be licensed under the GPL to maintain the competition. 52 In fact, open intellectual property can actually spur higher rates of innovation among non-open intellectual property holders through competition. A study tracing the history of the open source typesetting program TeX found that the simultaneous existence of commercial, closed source projects and open source projects benefited consumers because when one set of projects failed to meet a need in the market, the competing set of projects stepped in to meet the need. 53 Moreover, as OPAs remain open over time to all-comers and gain strength as they gain more network members, this is also likely to be considered procompetitive. 54

## OPAs in the absence of traditional patent licensing markets:

*p. 13*
benefits in the traditional market All of the previous hypotheticals assume a world in which a traditional patent licensing model remains viable, or in which the possibility of such a market remains viable. What about a market in which the OPA completely displaces traditional patent licensing? If the OPA achieves this position through conspiracy, that might still be punishable under traditional antitrust law. However, even without a traditional conspiracy, a market could develop in which the competitive benefits of joining an OPA are so strong that no rational market entrant would ever choose to employ a traditional patent licensing strategy. Such a market would lead to two questions: (1) would companies continue to innovate in the absence of a traditional licensing market? and (2) if the companies do decrease or stop producing new products, will this decrease innovation in the market?

*p. 13*
Turning to the first question: would companies still patent technology? Because OPAs draw their strength from the size of their networks, having patents is a key ingredient to OPA success. In addition, because OPAs also desire to protect their members from frivolous lawsuits by nonmembers, there is a strong incentive to file patents in order to block non-member attempts to claim patents for OPA innovations. In terms of the second question: assuming the OPA removes all incentive to patent, will companies continue to innovate? As open source software and now hardware companies have demonstrated, strong incentives exist to innovate using open licensing models; in fact, many companies that contribute to the GNU/Linux codebase are not remunerated directly for their expenditures, but recoup those costs through providing support to enterprises using the software. 55 Moreover, as Justice Stevens has noted in concurrence, "firms that innovate often capture long-term benefits from doing so, thanks to various first-mover advantages, including lock-ins, branding, and networking effects." 56 Is patent reform via private ordering anticompetitive? supported these theories. 57 In addition, other intellectual property law, such as trade secrets or trademark, may still protect innovations that are not protected by patent law. Thus, innovation is likely to continue even in the unlikely event that patenting rates decrease or slow down.

*p. 14*
Another benefit of providing free patents is that, by increasing the number of entrants into a market, the number of people and institutions dedicated to improving the technology in that market increases. The DOJ and the FTC have stated that "[r]estraints that encourage licensees to develop and market the licensed technology or that reduce the transaction costs of licensing the technology are more likely to be found reasonable." 58 In open source contexts, this benefit has long been considered an important part of the reason to open source a project. As early open source pioneer Eric S. Raymond put it, "[g]iven a large enough betatester and co-developer base, almost every problem will be characterized quickly and the fix obvious to someone" (or, in its most famous formulation, "given enough eyes, all bugs are shallow"). 59 The benefit of these improvements will then be available to OPA users for free, speeding the dissemination of new technology and benefiting consumers.

*p. 14*
Attracting new entrants may also steal mindshare away from rival enterprises. For instance, Google's smartphone operating system, Android, was able to quickly expand its market share in part because it was released under an open license and smartphone manufacturers and developers were able to access it for free. 60 By freely licensing its intellectual property, Google ensured that no other company (such as Apple or Nokia, the two dominant smartphone manufacturers of the time) would gain enough market power to hurt Google's core business of online advertising. Android is presently the dominant mobile phone operating system in the world, with over 80% of the market. See IDC, Smartphone OS Marketshare, 2016 Q2, http://www.idc.com/ prodserv/smartphone-os-market-share.jsp. While Google has faced antitrust scrutiny over its actions promoting Android, the allegations all involve Google's refusal to license its closed-source programs without bundling of other Google services, not over its open source codebase. See Mark Scott, Russian Authorities Rule Google Broke Antitrust Regulations, N.Y. TIMES (Sept. 14, 2015), http:// 3. OPAs in the absence of traditional licensing markets: benefits outside of the traditional market Traditional market benefits like those mentioned in the last section do not capture the full range of effects created by private ordering systems. Many actors create value, both benefits to social welfare and concrete monetary value, in response to non-monetary incentives. OPAs are well positioned to harness some of these motivations, with procompetitive effects.

*p. 15*
Because courts have long made clear that any social welfare created other than economic efficiency is irrelevant for the purposes of antitrust liability, 62 this discussion will focus on the economic benefits of ideologically motivated actors.

*p. 15*
As Judge Easterbrook wrote about software licensed under the GPL, "[t]he lack of money changing hands in open source licensing should not be presumed to mean that there is no economic consideration … . There are substantial benefits, including economic benefits[.]" 63 Wikipedia, for instance, was constructed out of volunteer labor and is largely given away for free; while difficult to value, estimates of its free market commercial value place it in the tens of billions of dollars. 64 The incentives to contribute such volunteer labor may be ideological ("I am benefitting my beliefs"), social ("I gain recognition from my peers"), or something else. However, the effect of harnessing these motivations may be to open up markets that did not exist and could not exist without an OPA.

*p. 15*
In fact, OPAs arguably solve or mitigate certain anticompetitive effects of traditional patent markets. Innovation does not require a patent system, and it especially does not require a patent system structured the way that the United States currently structures its patent system. Scholars have long recognized that the traditional patent licensing market suffers from market inefficiencies, and many have questioned if the money spent on patent licensing, litigation, and enforcement comes at the expense of other, more procompetitive business activities. As an alternative, some economists believe that "the most robust markets for ideas are those where ideas are free." 65 Prices of IP at any price above zero can have a radically chilling effect on potential entrants into the market for that IP. 66 Patent licensing markets are famously opaque. It is difficult for potential licensees to get information about whether patents will survive a challenge, whether other patents read onto their invention, and how much other licensees are paying for the same patent. 67 Patent holders have incentives to obfuscate, both by exaggerating the worth of their patents and by hiding the patents until it is too late for a potential licensee to redesign a product to avoid infringement. By offering everything for free, OPAs solve this problem. In addition, OPAs provide a way for patent holders to get some value in exchange for their patents (in the form of licenses and non-assertion guarantees from other DPL users) with transaction costs that are almost nonexistent. The U.S. Supreme Court has recognized that high transaction costs are a valid business justification for coordination of licensing when it found that music publishing license company BMI was not liable for antitrust violations for coordinating licenses across the entire music industry. 68 Economists have called the sensibility that drives certain people to exit markets on ideological grounds "repugnance." 69 Thus, for instance, people are resistant to using a traditional money-for-goods market to allocate human organs, so a well-designed market for human organs will not place a monetary value on the organs themselves. 70 Similarly, the appropriate scope of intellectual property rights, and in particular the potential for expansive intellectual property regimes to restrict freedom, is a moral issue to many people. Organized opposition to the patent right on moral grounds has a long history; it extends at least as far back as the anti-patent movement in 19th-century Germany, which opposed patents not merely on economic grounds but because awarding patents unfairly See id. at 828 ("[N]ot only is the cost of information replication low, but the demand curve for information goods becomes highly elastic at a zero price (and relatively inelastic at any positive price)"). benefitted inventors at the expense of the scientists whose work provided the basis for the inventions. 71 Today, many activists favor categorically ceasing patent protection for software patents. 72 Even those not morally opposed to the entire IP market structure may object to aggressive enforcement of IP rights. For instance, the record companies that sued end user copyright pirates suffered large amounts of bad press, 73 and a patent troll that sued podcasters for royalties wound up in a highly publicized clash with comic Adam Carolla. 74 If potential innovators who refuse to participate in a traditional patent licensing market can be enticed into innovating to join an OPA, the entire economy may realize large benefits-benefits that could outweigh those that the market would have generated under the traditional patent licensing model.

*p. 17*
Given the range of procompetitive benefits available to a user of an OPA, it is difficult to imagine that a court would find the implementation of an OPA violates antitrust law, even if the OPA operates to the exclusion of traditional patent licensing.

## CONCLUSION

*p. 17*
Private-ordering patent strategies, at least those implementing an open patent agreement, should prove robust in the face of antitrust challenges. While some embrace the patent licensing market as it is and attempt to work within it, open patent agreements offer an alternative vision of incentives for innovation. Fortunately, antitrust law is flexible enough to incorporate both perspectives, which gives patent holders and licensees the freedom to innovate not just in their products, but in how they structure their businesses. As the United States looks to reform its patent system, open patent agreements may prove to be important models for the legal system of the future.

## Footnotes

> 20 State Oil Co. v. Khan, 522 U.S. 3, 10 (1997) (internal quotation marks omitted) (quoting Fed. Trade Comm'n v. Ind. Fed'n of Dentists, 476 U.S. 447, 458-59 (1986)).

> 33DOJ & FTC, GUIDELINES, supra note 22, at 67. 34 See Texaco Inc. v. Dagher, 547 U.S. 1, 5 (2006).