This paper explores the application of EU Competition Law to the exploitation of human genome editing technology. Holders of key patents in the sector have applied different methods for disseminating the technology, such as different forms of licensing agreement and patent pools. It is found that that the competition rules are ill-suited to assess some of the licensing arrangements applied, which give rise to legal uncertainty. Accordingly, holders of patents on human genome editing technology may be discouraged to apply efficient methods for disseminating the technology. This may delay or obstruct some of the benefits the technology is supposed to deliver to the market, maker actors and consumers.
This article discusses potential competition law issues that may emerge as regards the exploitation of human genome editing technology. Ever since human genome editing technology was developed in the end of the past century, it has been a rapid developing technology. The technology has been described a bearing great potential for developing cures to diseases as the technology permits the modifying of cells in the human body. In particular, it is expected that human genome editing may be used to cure and treat diseases which so far have not been possible to treat effectively. While the technology promises great benefits to society, it is apparent that such technology raises a number of interesting legal issues related to ethics, public policy, ownership, dissemination and governance. Access to the technology and its exploitation, may under certain conditions raise competition law issues. In particular, human genome editing technology are protected by patent rights, meaning that access and use to the technology is controlled and “regulated” by those private parties that hold the relevant patents. As these entities enter into different forms of licensing arrangements, as e.g. patent pools, such collaborations may collide with different provisions in competition law. In addition, when discussing ground-breaking and valuable technology, there is always a risk that the holders of such technology may find themselves in a position of significant market power, which potentially opens up for compulsory licensing in individual cases under competition law. These competition law issues are discussed in this paper.
Importantly, the readers of this paper are expected to be mainly non-competition lawyers, but also to a limited extent competition lawyers that are interested in the application of the competition rules to the exploitation of human genome editing technology. For this reason, some basic elements that may be self-evident for most competition lawyers may deserve some additional explanation in this paper. At the same time, some elements of human genome editing technology, which are fascinating and may be important for the regulation of the sector with regards to concerns of ethical nature, public policy, public health and patent law, are not discussed in this paper. The focus of this paper is on the narrow and specific field of competition law and not general competition policy and regulation of the sector. Consequently, this paper does not focus on the general benefits and detriments of the technology as such, but rather on specific actions taken by companies within the sector that may fall within the scope of the competition rules. To a great extent, as will be explained below, the general benefits of the technology are unlikely to affect the competition law analysis and may only indirectly have an impact on the legality of what companies do in terms of regulating, granting and refusing access to technology on human genome technology. The main contribution of this paper is the discussion on how competition law may limit the behavior of holders of patent rights within this particular field of technology. Naturally, such limitations may have an impact on the dissemination of this important technology and its benefit to society.
A brief introduction to some key elements in the exploitation of the technology, which are relevant for competition law are presented in Section 2. In Section 3, an inventory of the potential competition law issues is carried out. Finally in Section 4, some preliminary conclusions are drawn.
2 The Exploitation of Human Genome Editing Technology
Although technology for editing genes and the human genome has existed at least since the 1990s, it was first with the breakthrough of the new editing tool CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) that it was made possible to edit genomes in a more precise and effective way.1 In addition, with the use of the Cas9-protein, it has created an enormous potential for the use of this technology (CRISPR-Cas9), as it is possible to make cuts at specific DNA sequences permitting the addition, removal or altering of DNA in the genome. For instance, a cell that is infected with HIV could potentially be altered as to remove parts of the genome that would reproduce the virus and in the long-term avoid that the infection develops into a full-scale AIDS. As the technology can be used to alter the genome in all living organisms, it also means that it may be used to alter genomes in plants or animals, e.g., by introducing parts in the genome that would make it resistant to a particular type of disease.2 Obviously, the technology as such has great potential, in particular for the development of more effective treatment to human diseases and the promotion of public health, but also for improvements and the protection of other living organisms such as crops and animals used for human consumption.
While the technology and its potential uses is a fascinating topic in itself, it is not so interesting from a competition law perspective. The interesting part from a competition law perspective is the competitive situation as regards access to the technology in question, as well as the exploitation of that technology in markets for products and services. As the technology is held by private parties or state actors (like some universities) acting as private actors (by e.g. transferring the patent rights to a company under their control), access to the technology is mainly governed by the holders of the technology. Naturally, there are other bodies of rules that may affect access to the technology or its use. Regulation on what kind of research that is permitted or not within a particular jurisdiction will affect the further development of technology (follow-on innovation), which is a particular kind of competition between market actors. Regulation on legal and illegal therapeutic uses will impact to what an extent there will be a market for a particular application of the technology. Moreover, the rules in patent law, which may be different in between different jurisdictions, may prohibit the patenting of certain technologies or stipulate exemptions to the exclusive right for particular uses of the technology. For instance, in certain jurisdictions there may be a wider research exemption which permits researchers to further develop the technology without permission from the owner of a patent. Importantly, an analysis under the competition rules will take the legal framework provided by other bodies of rules as granted and as part of the market conditions in a specific case. However, it must also be noted that just because a particular behavior is tolerated by other bodies of rules, it does not mean that the behavior is immune from the application of competition law.3
From what may be read out from the literature at the moment, there are few elements regarding the competitive situation related to the technology itself as well as to the access to the technology that are of particular interest from a competition law perspective. Firstly, like in many other markets which are driven by R&D there is a presence of some key patent rights on the basic technology as well as hundreds of patent rights which constitute different developments of the basic technology. Importantly, the necessity to receive licenses under certain patent rights to use the basic technology means that a few patent holders may be in a position of so-called market power as the exclusive right grants them a legal monopoly to use the technology. This may trigger the application of competition law as discussed more specifically below (Section 3.5).
Moreover, as in many other fields where different actors are racing for patenting new technology, several actors have been involved in patent disputes regarding the CRISPR-Cas9 technology.4 Although the details of these disputes are irrelevant for the purpose of this paper, it follows that the litigation has created uncertainty about issue of ownership, which also has repercussion for those companies that would like to access and use the technology. As patent rights to both the basic technology as well as to more specific developments of that technology are held by different market actors, there may be a need for several patent holders to collaborate through agreements either for the purpose of using the technology or licensing the technology to interested third parties. Such collaborations, depending on the design of the relevant agreements, may also trigger competition law as they may reduce competition between the parties or between the patent holders and third parties. For instance, one such type of collaboration that has been discussed is patent pools, whereby several patent holders pool their technologies to license to third parties. The MPEG-LA launched an initiative in 2020 to create a patent pool consisting of CRISPR-Cas9, even though not all key actors have joined in this initiative.5
Furthermore, there are also some few details concerning existing and planned collaborations as regards the CRISPR-Cas9 technology that are interesting from a competition law perspective. To begin with, it seems as some universities use a licensing arrangement referred to as surrogate licensing. With surrogate licensing it is meant that the commercialization and licensing is outsourced to a private party (surrogate licensor).6 For instance, one the technology holders of the patents of CRISPR-Cas9 (University of California) has licensed the technology to a third party that both develops and licenses the technology to others. In other words, the institute has completely given up control of the technology to its licensee. Such licensing arrangements seem primarily to have been concluded in order to avoid certain obligations imposed on universities, which is not necessarily a relevant factor for the competition law analysis.7 However, as discussed below in Section 3.3, the use of surrogate licenses may have an impact on which specific rules that apply under EU Competition law, which makes the use of this type of licensing interesting.
Additionally, it has been mentioned above that the CRISPR-Cas9 system could have a variety of different uses. Technology holders may therefore limit the use of the technology for particular purposes, through so-called field-of-use restrictions.8 For instance, one of the patent holders on CRISPR-Cas9 (the Broad institute) has retained the licensing for non-commercial non-human therapy uses, while a surrogate licensor licenses the technology for commercial use. The surrogate licensor has in turn subsequently granted a specific license for the use of the technology in plants. Field-of-use limitations are typical license restrictions. They may be used in order to protect certain markets where either the licensor or other licensees are active. Technology holders may have legitimate reasons for protecting their own commercial activities or those of other licensees in order to create incentives to invest in the development of the technology within a particular field-of-use. A field-of-use can also be seen as reflecting the specialization of a firm’s commercial activities. However, field-of-use restrictions may also be used to divide markets between market actors, meaning that each actor may face none or very little competition from other market actors within a specific product market, which make such restrictions interesting for a competition law assessment.9
Importantly, in the context of CRISPR-Cas9, there is also a discussion on so-called ethical licenses. The meaning of an ethical license is that the licensee is prohibited to use the licensed technology for specific purposes that are deemed as unethical.10 The (potential) use of ethical licenses are interesting because they constitute a way for technology holders to introduce conditions that would not otherwise follow from national legislation where a licensee is active. For instance, if national legislation would not prohibit the use of CRISPR-Cas9 in research to manipulate the genome in human embryos, the technology holders could impose such limitations (under the condition that they have patent protection in the country where the technology is used). Accordingly, a group of technology holders could use ethical licenses as a form of self-regulation within specific markets or for certain types of research. Under competition law, limitations through ethical licenses constitute or are comparable to a field-of-use restriction, which make them interesting to discuss below.
Finally, it is also necessary to mention that licenses may be exclusive or non-exclusive. This is a typical feature of licensing agreements in general and seems also to apply in licensing of genome editing technology.11 As a licensee may have to invest heavily into the further development of the used technology, it is not uncommon that a licensee may request and will be granted an exclusive license in order to have sufficient incentives to develop the technology. Importantly, without exclusivity, some market actors may not be willing to risk heavy investments into the further development or commercialization of new technology. Exclusivity may concern either particular fields-of-use or territories. As mentioned above, certain surrogate licensees have received an exclusive license for a particular purpose, such as the commercial use of CRISPR-Cas9 technology. From a competition law perspective, exclusive licenses may be problematic if they result in that only one company will be able to compete in a particular market. The licensing of several exclusive licenses may also be used to prevent competition in between licensees and thus the division of markets. Accordingly, exclusivity is also an interesting feature that is relevant to discuss.
3 Potential Competition Law Issues
There are several unknows about the governance and the dissemination of the human genome editing technology that makes a more specific discussion of the potential competition law issues somewhat speculative. Accordingly, this section purports to make an inventory of possible issues under competition law, rather than giving straightforward answers about the legality of particular governance arrangements. In such a way, possible weaknesses with the current competition law regime may be identified. Considering that the competition law regime for R&D cooperation agreements is up for review by the Commission,12 and that a review of the rules on technology transfer (patent and know-how licensing for the purposes of producing particular products or services) is likely to occur in the coming years, the inventory below may raise interesting issues for further discussion.
As mentioned above, it seems as the dissemination of human genome editing technology at the moment is being carried out through a variety of private governance methods, including patent pools or ordinary licensing agreements. It seems as licensing to other research institutes and for non-profit purposes is being done through non-exclusive licenses. Some of the commercial exploitation, including further research, is being carried out through exclusive licenses. In addition, there is a discussion about the use of licensing agreements for ethical uses. Even though ethical licenses are probably not very problematic according to competition law, they will be briefly touched upon below. Moreover, under competition law there is a possibility to force access to patented protected technology to prevent patent holders to monopolize existing and potential markets for product and services. This is also addressed below.
Importantly, the analysis below is based on accounts of licensing arrangements given by other academics. These accounts are naturally too superficial to give a good factual basis for a competition law analysis. The arrangements discussed below should therefore be seen as examples of possible licensing arrangements.
3.1 An Overview of the Relevant Competition Rules
Although somewhat simplified, it could be said that competition law mainly deals with problems related to market power. With market power it is meant that a market actor or a group of market actors to a certain extent may determine market conditions, such as price and output, for particular products or services. By contrast, under conditions of effective competition undertakings would normally have to adapt themselves to market conditions in order to survive on the market, being so-called price takers. A position of market power, in various degrees, may be reached through cooperation between market actors, in particular through agreements between competitors, but also between non-competitors, which may limit competition to the extent that the cooperating parties may influence market conditions in terms of price, quality and innovation. A position of market power for individual market actors may also emerge as natural development in the market. However, such market power may be exploited to restrict competition from other market actors, but also to impose unfair conditions on customer and consumers. Competition law purports to hinder the emergence and exploitation of market power through cooperation as well as prevent exploitation of market power unilaterally by individual market actors. In addition, in EU Competition Law there is the additional goal of promoting the internal market. This affects the application of competition law to both cooperation and unilateral conduct that may reduce or obstruct trade between Member States, as this is viewed as contravening the goal of integrating Member States’ national markets into one internal market.
The main provisions in EU Competition Law that apply to the actions and behaviors of private parties are Articles 101 and 102 TFEU. Article 101 TFEU prohibits anti-competitive agreements that restrict competition and affect trade between Member States under the prohibition under Article 101(1) TFEU and which do not meet the requirements for the exemption under Article 101(3) TFEU. The provision aims in particular to capture agreements that result in increased prices, the limitation of output, the sharing of markets and the hindrance of competition through R&D and the exploitation of technology.
Article 101(3) TFEU exempts agreements that result in an economic benefit, consumer benefit through restrictions of competition that are indispensable for the beneficial effects of the agreement while not eliminating competition for a substantial part of the product/services in question. As there is not much case law on the application on Article 101(3) TFEU, which is still surrounded by uncertainty and vagueness, the so-called block exemptions regulations adopted by the Commission are of utmost importance. Block exemptions refer to those regulations that pinpoints more specific requirements for particular categories of agreements, such as vertical distribution agreements, R&D agreements and technology transfer agreements, and that are exempted on ‘block’. The block exemptions therefore constitute an expression of the application of the requirements in Article 101(3) TFEU to specific categories of agreements.
In addition, the Commission also adopt soft law instruments, such as guidelines and notices, that accompany particular block exemptions by explaining the rules in those regulations. Moreover, they give a general account of how the Commission views and analyzes specific category of agreements and contract clauses under Article 101(1) and (3) TFEU. While the Commission’s guidelines constitute soft law, meaning that those rules are not binding on the Union Courts, national courts and the national competition authorities, they may be binding upon the Commission. In practice, the Commission’s soft law provide with much valuable guidance by providing an accurate account of competition law as it stands, in particular regarding areas where the case law is scarce, as e.g. technology transfer agreements. In the sections below there are several references to both block exemptions and the Commission guidelines.
Article 102 TFEU deals with unilateral conduct. The provision prohibits the abusive behavior by a company with a certain degree of market power, dominance, that allows it to a certain extent behave independently from its competitors, its customers and ultimately the end-consumers. In a competitive market each undertaking has to a adapt itself to market conditions such as price, quality of the products or sales methods in order to remain on the market. However, a dominant undertaking will have the power to influence or set market conditions to a certain extent. The problems that may emerge with a dominant undertaking is that it may engage in three types of abusive behavior that may be detrimental to the market. First, it may exploit its market power by extracting benefits from customers that it could probably not extract in the absence of market power, e.g. by imposing a supra-competitive price. Secondly, it may also use its market power to exclude competitors (exclusionary abuse) in an effort to strengthen or maintain its market position or to expand its market power to related markets. Thirdly, it may use its market power to harm integration in the single market, normally with the purpose of protecting itself from parallel trade or to extract higher profit margins in different Member States. Finally, while Article 102 TFEU does not include a rule that gives possibilities for exemption, it follows from the Court’s case law that is possible to justify prima facie abusive behavior with so called objective justifications or by demonstrating that the abusive conduct results in efficiencies that outweigh the negative effects on competition.
3.2 The Applicable Block Exemptions in EU Competition Law
One important but tricky issue in the competition law regime is the application of the block exemption regulations to different licensing agreements. Even though Article 101(1) and (3) TFEU may always be applied, there is a considerable uncertainty in the interpretation and application of these rules as case law is relatively scarce and the Commission’s soft law is not sector specific. For a company to be on the safe side it is best, if possible, to adapt licensing and cooperation agreements as to make them comply with the block exemption regulations. Naturally, as the design of licensing agreements usually have to consider idiosyncrasies of a particular sector, the particular technology that is subject to license, and the risks with collaboration caused by e.g. free riding or uncertainties, this is not always possible to do. The benefits with being block exempted is that even if an agreement would ultimately be found problematic in an individual case, exemption under one of the block exemptions regulations would afford protection from sanctions under competition law until the Commission withdraws the block exemption from the specific agreement.
It seems as the owners of the CRISPR-Cas9 technology are engaged in licensing for both further R&D and the commercial application the technology. Importantly, both the block exemptions on R&D cooperation and technology transfer may prima facie be applied to licensing agreements. Regulation 316/2014 on the application of Article 101(3) TFEU (TTBER) applies to so-called technology transfer agreements. Technology transfer is defined as the licensing of technology rights for the production of the contract products.13 Regulation 1217/2010 on the application of Article 101(3) TFEU to R&D agreements (RBER),14 also applies to licensing of technology rights if they occur within the context of a R&D collaboration. R&D agreements under the RBER includes, inter alia, the licensing of technology rights shared for the purpose of conducting new joint research and exploitation,15 and the licensing of technology as a form of exploitation if the licensing is pursuant to R&D conducted by the parties under the agreement or pursuant to a previous R&D agreement.16
The distinction between the scope of one block exemption and the other is not always easy to make. The crucial element in the definition of technology transfer in the TTBER is that the licensing is made for the manufacture and sales of the contract products on the market. With other words, the focus of the block exemption is the licensing of products/services that are supposed to be manufactured and placed on the market by the licensee. There may be some R&D activity related to such an agreement, as e.g. the development of a manufacturing process for the contracted products (which are protected by the licensed technology rights). However, if the primary purpose of the license is to conduct R&D, the agreement will fall outside the scope of the TTBER. According to the Commission Guidelines on Technology Transfer (TT Guidelines), it is required that the contract products have been identified for the TTBER to apply.17 As regards licensing agreements with the purpose of manufacturing and selling the contract products, they may in theory fall both under the RBER and the TTBER. Such a licensing agreement falls, prima facie, under the scope of the TTBER. However, if the licensing agreement concerns technology rights that are the output of an R&D collaboration between the parties (e.g. through a previous R&D agreement), the RBER applies.18 Importantly, the TTBER does not apply to agreements which are covered by the RBER.19
The distinction between the TTBER and RBER is important as the latter has more generous rules. In particular, when the research agreement is concluded between non-competitors, the RBER exempts the agreement between the parties during the research development phase and up to seven years in the exploitation phase from that the contracts products are placed in the market.20 Importantly, as long as the parties are non-competitors, there is no market share ceiling during this time. After the seven years period, the agreement is exempted as long as the parties cumulative do not exceed a market share ceiling of 25%.21 By contrast, under the TTBER, licensing between non-competitors is exempted when the parties do not exceed market shares of 30% in their respective markets.22 In addition, also the other requirements under the block exemptions are more generous under the RBER than the rules in the TTBER. For instance, the RBER permits non-competition clauses (meaning that the parties to the agreement may be prohibited from selling competing products or technology) during the time the parties jointly exploit the results.23 By contrast, the TTBER exclude such clauses from the block exemption and require an individual assessment under Article 101 TFEU, while the rest of the agreement may still be block exempted under the TTBER.24 Also as regards cooperation between competitors, the RBER is more generous than the rules in the TTBER. The RBER applies a 25% market share threshold, while the TTBER applies a market share threshold of 20%.25 The RBER also permits some restrictions between the competing parties that would not be exempted under the TTBER. For instance, as the RBER permits the joint exploitation of the results of R&D, such as the licensing of the researched technology to third parties, a uniform royalty can be charged for such licenses.26 Any coordination on prices would however be excluded under the TTBER.27
It follows that RBER is more beneficial than the rules in the TTBER. It is also likely that at an early stage licensing regarding CRISPR-Cas9 technology would primarily concern agreements for the purpose of the further R&D, rather than the manufacturing of specific products and services. The more lenient treatment of licenses for the purpose of R&D follows the logic in competition law that the farther away from sales of specific products and services on the market, the less problematic an agreement is likely to be for competition. For licensing that occurs “closer” to a market, like a technology transfer agreement, where the contracts products that are going to be manufactured and sold by the licensee are already determined, the conditions for block exempting such agreements become stricter. The lenient treatment of licensing at an early research stage under the RBER may be important in the development of CRISPR-Cas9 technology. It is likely that there may be market actors that will hold key patents and perhaps even the only patents within a specific field-of use. The RBER will still apply as long as licensing does not occur with competitors, which somewhat simplified refers to licensees that are in hold of competing technology.
It should also be noted that where the RBER and TTBER overlap, the legal regime has a more favorable view on long-term R&D cooperation between two parties. Parties that have engaged in a previous R&D agreement (that falls within the RBER) can also benefit from the more lenient rules in the RBER (instead of the TTBER) at a later stage when the technology can be exploited for sales of specific products and services.
Obviously, the main issue with both the RBER (for licensing between competitors) and the TTBER is the application of market share thresholds. As stated above, with the development of new technology there is a risk that market shares may be high for certain actors that have key patents on CRISPR-Cas9 technology. Accordingly, there may always exist a risk that certain technology holders will not be covered by the block exemptions when licensing to other market actors. However, from a competition law perspective it is inevitable that an individual assessment must be made of agreements when there is market power. As discussed below, the main problem with the current legal regime is probably not an overly strict assessment of licensing agreements under the RBER or the TTBER because of market share thresholds, but rather other issues related to the scope of the block exemption regulations.
3.3 Surrogate, Exclusive and Ethical Licenses under EU Competition Law
As stated above, some owners of the key patents in the CRISPR-Cas9 technology have concluded surrogate licenses with partners (surrogate licensors), which in turn have concluded exclusive licenses limited to a certain field-of-use with third parties (third-party license). In addition, in the context of CRISPR-Cas9 technology there is also a discussion on the use of ethical licenses with the aim to prevent that licensees engage in research or application of the technology that raise problems with ethical concerns.
Importantly, as concerns surrogate licenses, it is first interesting to discuss the relation between the academic institutions and those partners that have been granted an exclusive license for the exploitation of the CRISPR-Cas9 technology. Importantly, if the academic institutions are shareholders or owners of the partner that has received a surrogate license, such arrangement may fall outside of Article 101(1) TFEU. According to the doctrine of an economic unit, agreements with undertakings that are controlled by the licensor are not viewed as agreements between undertakings. Accordingly, Article 101(1) TFEU would not capture such arrangements irrespectively of the conditions in such licenses.28
If surrogate licenses are concluded with independent undertakings, or if licenses are concluded between the surrogate licensor and independent third parties (third-party licensees), such arrangements could fall within Article 101(1) TFEU. Exactly which legal regime that would regulate such licenses, depends on the purpose of the license. As discussed above, if the surrogate licensor only exploits the technology by further R&D (including the use of a research tool), it could fall within the scope of the RBER. However, it is important to note that an arrangement whereby only one party would conduct R&D does not fall within the RBER. The RBER only captures “joint” R&D where there is a collaboration between two parties.29 Granting a license for the purpose of R&D to another undertaking is not necessarily captured by the definitions of joint R&D. While the RBER covers paid-for research, defined as R&D carried out by one party and financed by another,30 the task imposed on a licensee to license to third parties is not covered by the term R&D. In addition, the RBER also requires that the parties have full access to the results and may also require access to pre-existing know-how that the parties have “brought into” the R&D collaboration.31 These requirements are too technical and complex to elaborate further on here, but it may be said that it seems unlikely that the current licensing arrangements on CRISPR-Cas9 technology would necessarily meet these requirements. Accordingly, licensing agreements where one party is merely given the permission to exploit patent rights to conduct research without involving any elaborated cooperation between the parties are likely to fail the preconditions for being covered by the RBER. In addition, such licenses with the purpose of R&D would also fall outside the TTBER. As regards R&D that is carried out by the surrogate licensor independently and that is not financed by the other party would also probably fall outside both the RBER and TTBER.32 If the license however is used to manufacture and sell specific contract products, it falls under TTBER.
It follows that rules in the RBER are complex and may give rise to uncertainty of the coverage of R&D licensing agreements. This is also a theme that has been raised in the Commission’s review of the RBER and the Horizontal Guidelines. However, that discussion has mainly concerned the requirements of granting all parties to an R&D agreement full access to the results as well as pre-existing know-how.33 The review does not seem to focus on the requirement of “joint” research which may also be problematic for surrogate license arrangements, including third-party licenses. Naturally, the current legal regime under the RBER and TTBER may raise problems of legal certainty as regards the licensing agreements that are concluded on CRISPR-Cas9 technology. In the worst case scenario, an ordinary licensing agreement for the purpose of R&D that does not entail any actual collaboration between the parties (like R&D conducted by a joint research team) may have to be assessed directly under Article 101 TFEU. On the other hand, if the license between an academic institution and surrogate licensor are non-competitors, which would seem likely considering that academic institutions are involved, it is unlikely that such a license would raise any competition concerns. An academic institution would in such cases not be involved in licensing to third parties of any technology in competition with the technology licensed by the surrogate licensor. Even if the surrogate licensor has a high market share, the license would probably be unproblematic as no competition is restricted between the parties. However, if the academic institution would continuously license technology exclusively to the surrogate licensor, which reinforces or maintains a position of market power of the latter, it could infringe Article 102 TFEU.34
The assessment may be more difficult as regards licenses between the surrogate licensor and licensees. As mentioned in the literature on CRISPR-Cas9 technology, it seems as surrogate licensors may engage in both further R&D of the technology as well as exploitation by licensing to third parties. In such cases, there is a greater likelihood that the surrogate licensor may constitute a competitor to the licensee (at the time of licensing or later), which can make the assessment under Article 101(1) TFEU more difficult, as agreements between competitors more easily can restrict competition. In addition, as it seems as surrogate licensors may grant exclusive licenses limited to particular fields-of use (see above, section 2), a network of such licenses may be seen as dividing market between licensees, where each licensee is granted exclusivity within a particular market. If those licensees could potentially be active in each other’s exclusive field-of use (and thereby compete), an assessment outside the block exemption regulations could be problematic under competition law.
In situations where a surrogate license or a third party-license fall within the RBER, the rules are fairly generous for granting exclusive licenses. Importantly, EU block exemptions are obsessively focused on territorial exclusivity. This is motivated by the protection of the single market (see above, Section 3.1) and the fact that territorial protection has the very purpose to restrict trade. However, the little information on the forms of governance of CRISPR-Cas9 technology indicates that exclusivity normally refers to field-of-use restrictions, for instance that a license is limited to the use for producing a particular variant of a plant. To begin with, the RBER recognizes this type of licensing restriction as form of specialization which falls under the RBER as long as the R&D collaboration is classified as joint R&D under the regulation.35 Such licenses may also be exempted as they are not classified as hardcore restrictions,36 as long as the license respect the market share thresholds discussed above (see above, Section 3.2).
In case a license (probably a third-party license) falls under the TTBER, the rules are more or less the same. In agreement between competitors, a field-of-use restriction is permitted under the condition that no field of use is imposed on the licensor and that the market share threshold of 20% is not exceeded.37 In agreements between non-competitors, field of use restrictions imposed on both licensor and licensee are tolerated under the market share threshold of 30%.38
It is also pertinent to address the issue of so-called ethical licenses. It is discussed in the literature that ethical licenses may be used in order for patent holder to control the use of the technology by the licensee. Importantly, ethical licenses could be used by key patent holders to impose a form of self-regulation in the sector with the purpose of promoting ethical or other societal goals which are not directly connected to the economic aspects of exploiting the technology. Naturally, ethical licenses can be designed in different ways and with different restrictions which may or may not be problematic from EU Competition Law perspective. For instance, ethical licensing may include the obligation on the licensee to grant farmers rights to save and resew seeds for the next coming year.39 One form of ethical licensing that is interesting for competition law are licenses that prevent the use human genome editing technology, like CRISPR-Cas9, for certain uses that may raise ethical concerns,40 such as the extreme example of Lulu and Nana, the genome-edited twins.41 The key aspect from a competition law perspective is that such licenses permit companies to use the technology for particular uses while excluding the same companies from entering a particular market. Obviously, in theory, such licensing can be used for allocating markets between different parties, which restrict competition. Although it is not completely certain, such an “ethical” restriction imposed through a license would amount to a technical field-of-use restriction.42 Currently, it seems likely that an ethical license would primarily be used in a R&D collaboration as the restriction is aimed at preventing a certain type of research rather than the application of already existing technology. Accordingly, such ethical licenses would be assessed in the same manner as the field-of-use restrictions discussed above.
As discussed above, a potential problem with field-of-use restrictions is that they can be used to divide up markets between different licensees, which could restrict competition. In particular, if such licenses would not fall under the block exemption regulations and involve a larger network of licensees which are granted a particular exclusive product market, they could potentially be struck down under the competition rules. However, it seems unlikely that the imposition of ethical restrictions through field-of-use restrictions could result in such a division of market. If the licensor genuinely prevents all licensees for applying the technology for some unethical uses, they would not result in carving out particular markets where only one licensee would be active. If the patent holder is consistent by excluding all licensees from a certain field-of-use, there should not be any problem under competition law.
Would an ethical license nonetheless be found to restrict competition, it should be noted that it may be difficult to justify such a restriction on ethical grounds. The exemption under Article 101(3) TFEU does not give much room for “soft” goals unconnected to the improvement of efficiency (by reducing costs or increasing quality of the product and services affected by the agreements entered into by the parties). And while there is case-law that supports that private parties may enter agreements that constitutes a form of self-regulation, such as setting rules of conduct for lawyers by a bar association or rules of the game in sports, it seems unlikely that this would apply to a self-regulation on the use of CRISPR-Cas9 technology.43 Importantly, those cases have concerned the regulation of particular activities and professions, where the state has left the regulation of those activities to the sector itself. A self-regulation regarding CRISPR-Cas9 technology would firstly concern a much broader area as technology can be used for multitude of activities and markets. Moreover, several aspects of the technology are already directly or indirectly regulated by patent law, pharmaceutical law etc. At first glance, it appears therefore as it would be difficult to justify ethical field-of-use restrictions once a restriction of competition has been established in an individual case.
3.4 Patent Pooling of CRISPR-Cas9 Technologies
As mentioned above, some of the key patents protecting CRISPR-Cas9 technologies have been gathered in a patent pool run by the MPEG-LA.44 It is initially important to underline that patent pools may be both good and bad from a competition law perspective. In the past, patent pools were controversial for their similarity to the most serious type of agreements restricting competition, so called cartel agreements. A cartel agreement exists when competitors agree to stop competing with each other, e.g., by setting a uniform price for their products or dividing up markets by categories of products or by territory. The pooling of patent rights covering competing technology which is sold to a uniform price does not differ from a cartel agreement. In addition, the pooling of all technology necessary for the production of particular products or services can also grant market power to the collective of patent holders that are members of a patent pool. Such patent holders could therefore impose unfair conditions or restrict competition when entering into agreements with users of the technology. Patent pools may however also be highly beneficial. By accumulating all necessary technology, it reduces the costs for users that only need to turn to one “seller” instead of having to negotiate several licensing agreements, which may create a “bottleneck” if a particular patent holder would be unwilling to license. If the patent pool has on beforehand set a royalty (or the price) that users have to pay for a license, there is also no risk that a licensee may be “blackmailed” to pay exorbitant royalties. Patent pools may also reduce costs for patent holders to supervise potential infringers of their patent rights as well as costs of enforcement. Patent pools that license the pooled technology to all interested parties, without discrimination, may be an efficient way of disseminating technology.
Importantly, patent pools, including licenses from the patent pool to third party users, do not fall within the RBER or the TTBER. A patent pool is defined as an arrangement whereby two or more owners of patent rights (patent pool members) pool their patents for the purpose of licensing in between the patent pool members and/or to third parties (third-party users).45 The patent pool as such can be organized in different ways. Patents may be licensed to a third party that in turn licenses to third-party users, or the patents are licensed to one of the patent pool members, which licenses to third-party users. The focus of the RBER lies on the R&D cooperation between two or more undertakings. Licensing of technology that constitutes the output of such collaboration, so-called exploitation of the results, falls within the RBER. The mere gathering together of a package of technology for the sole purpose of licensing to third parties thus falls outside the scope of the RBER.
As regards the TTBER, the focus lies on the licensing of technology for manufacturing and sales, which at first blush may give the impression of covering patent pools as the pooled technologies are licenses for the purpose of manufacture and sales of contract products. However, the agreements that encompasses a sharing of technology between the owners of the technology included in the patent pool do not have the purpose of manufacturing and sales.46 In case the patent pool is created by the licensing to a third party that in turn licenses to third-party users, such an arrangement would also not have the purpose of manufacturing and sales. By contrast, a third-party license from the patent pool to interested users may have the purpose of manufacturing and sales. In such cases, the enabling regulation that gives the power to the Commission to adopt the TTBER limits the exemption of such licenses to be concluded between two parties.47 A third party license is seen as involving not only the pool, but also its owners. Thus, even patent pools with only two members, would be classified as an agreement between at least three parties.48 Accordingly, patent pool arrangements will always fall outside both the RBER and TTBER.
While patent pools and third-party licenses of the pooled technology do not benefit from the legal certainty of the RBER and TTBER, the TT Guidelines provide a ‘safe harbor’. With a safe harbor it is meant that when pools meet the conditions stated in the TT Guidelines, they are presumed to fall outside Article 101(1) TFEU, irrespective of the parties’ market positions.49 Patent pools are deemed to fall within the safe harbor when the following conditions are met: participation is open to all interested technology right owners; there are sufficient safeguards to only include essential patents in the pool; there are safeguards to limit exchange of sensitive information to what is necessary for the creation and the operation of the pool; the patent pool has non-exclusive licenses on the essential patents; licenses are given on FRAND (Fair, Reasonable and Non-Discriminatory terms) to interested users; parties are free to challenge the validity and the essentiality of the pooled patents; patent pool member and third-party users are free to develop competing products and technology.50
The TT Guidelines state that there is no inherent link between patent pools and standardization.51 Nonetheless, it seems as the safe harbor in the TT Guidelines has been designed in the light of arrangements concerning technical standards, which are typically applied in the sectors of telecom and electronics. This is illustrated, in particular, by the distinction between essential and non-essential patents, which is one relatively easy to make in technical standards as these are to some extent defined by the technology protected by the patent rights. The situation seems to be somewhat different in the biotech industry where standards cannot be “created” in the same fashion as in areas such as telecommunications.52 Moreover, a patent pool on CRISPR-Cas9 technology may include technology for a variety of uses, which are therefore not “essential.” Patent pools could also be used to resolve patent litigation issues by including those patent rights that are subject to litigation. Such an approach may facilitate for users that will not have to run the risk of infringing the patent of a third party even though they have already licensed the “necessary” technology from a patent holder which is involved in patent litigation. The reasons for creating a patent pool on CRISPR-Cas9 technologies may simply be different from the rationale behind patent pools that are connected to standardization. The problem is that such patent pools may fall afoul of the competition rules. It has been argued that a problem with pools including non-essential patent results is that they will not resolve the bottleneck issues that pools are supposed to resolve.53 Accordingly, such patent pools may be seen as more “suspicious” by competition authorities. From a practical perspective, a pool that includes non-essential patents may not benefit from the safe harbor described above.54 As a consequence, because of the uncertain status of patent pools under the competition rules, key patent holders may be discouraged from entering such arrangements. Importantly, as patent pools can be an efficient way of managing and disseminating key technology to a large group of potential users and for a variety of uses, competition law may delay or obstruct dissemination of technology that would be good for the market, market actors and ultimately the consumers.
3.5 Refusal to License
While the discussion above has concerned Article 101(1) TFEU and licensing, Article 102 TFEU may also be relevant for owners of the CRISPR-Cas9 technology. It has been reported that owners of technology have used exclusive licenses to certain undertakings for the commercial use of CRISPR-Cas9 technology. These licensees are in charge of either developing the technologies and/or sublicensing to third parties. The licensees have replaced the academic institutes with the task to commercially exploit the technology through the so-called surrogate licensing.55 These licensees have subsequently licensed the technology to third parties for either therapeutic use or use in plant technology. It has also been identified as potential problem that the licensees, which are involved in both further developing and applying the technology, as well as licensing, may not have an incentive to license to certain third parties, as they may constitute a competitive threat. Considering that the technology is fairly new and that there may not be viable substitutes, there is a risk that the holder of such technology may be, at least temporarily be in a dominant position. Accordingly, if such a holder of key technology refuses to license to third parties, this may raise the issue of whether such a refusal constitutes an abuse under Article 102 TFEU. Although the assessment of whether a company is in a dominant position is crucial, it may be difficult with the limited amount of information available to make a determination of dominance at the moment. Below, the discussion will therefore be limited to the issue of abuse assuming that a holder of technology is found to be dominant.
Under Article 102 TFEU, refusal to supply goods and services may be found abusive in particular when such a refusal may result in that the dominant undertaking eliminates competition on an adjacent market. Refusal to license intellectual property rights (IPRs) constitutes a ‘branch’ of the case law on refusal to supply. Because of the importance of IPRs for competition (through innovative activities) the Court has held a somewhat high threshold to find abuse in cases regarding refusal to license.
Refusal to license was first established as abuse in Magill.56 The case concerned the refusal to license copyrights from three TV-broadcasters to a company that wanted to put together TV-listings of programs into a weekly magazine. Oddly, although those listings were hardly an expression of original and creative work, which is the protected subject-matter under copyright, those listings were protected under national copyright law. The Court held that the refusal to license constituted an abuse. It found that the refusal had prevented the emergence of new product; reserved an adjacent market to the dominant company by eliminating competition in the market; there were no objective justifications for the refusal.57 In Bronner, the Court held that all refusal to supply cases also require the demonstration that the input product/services must be indispensable before an abuse can be established.58 The confirmation of the indispensability requirement in refusal to licenses case follows from IMS Health.59 Accordingly, a license must be considered as indispensable to enter the adjacent market for the product or service that incorporates the licensed technology.
The refusal to license cases, in essence, make a balance between the interest of maintaining competition and the protection of the interests under IPRs. Importantly, IPRs permit right holders to deny access to the use of an invention, artistic works, trademark, etc. The possibility to refuse a license constitute the very essence of IPRs. While it follows from Magill, IMS Health and Microsoft that under exceptional circumstances the interests of competition will triumph the protection afforded under IPRs, the threshold set by the case law is supposed to be relatively high.
In particular, the indispensability requirement normally sets a high threshold for finding an abuse in refusal to supply cases, although this may be different for IPR protected inputs as discussed below. With indispensability it is meant that the company requiring access must show that it would not be possible to enter the (adjacent) market without a license from dominant undertaking. The requesting company must not to be able to get supplies from other undertakings and should not be able to produce the input on its own. It is not sufficient that without a license it would become less economically viable for the requesting company to offer the goods/services in the adjacent market. The fact that it would be less economically viable for the requesting company to produce the input because it is an undertaking with less turnover than the dominant undertaking does not amount to indispensability.60 Thus, indispensability requires that a company which is comparable the dominant undertaking would not be able to reproduce the requested input.
In Commercial Solvents the Court found that alternative methods of production which could have afforded the requesting company an alternative source of supply could not be pursued in an industrial scale.61 By contrast, in Bronner, the Court hinted that there were no barriers, legal or technical, for the requesting company to establish its distribution network (of magazines), even though the establishment and operating of such a network could have been more costly than getting access to the network of the dominant undertaking.62
As regards indispensability and IPRs, it follows that the license in Magill was de facto indispensable. Obviously, if the content of the TV-listings were protected by copyright, it would not be possible to produce an alternative input as the requesting company needed an exact copy of copyrighted materials. However, the situation in Magill could be viewed as an extreme case. It would be unusual that a piece of text would constitute an input that could not be substituted with something else. However, IMS Health also concerned the request for copyright related rights. The case concerned the use of a database structure which had been created by the company NDC Health. The database structure had been developed together with the dominant undertaking’s customers. Some employees left NDC Health and created a competing service based on a database structure, which was more or less the same as the one offered by NDC Health and probably infringed the IPR protection on database structure. The case triggered litigation between the parties regarding the potential infringement of IPRs as well as competition law issues. Ultimately, the competition law case reached the Court through a preliminary ruling procedure. Assuming that there was a breach of NDC Health’s IPRs, the Court found that the license was indispensable. As the database structure had been developed in cooperation with the customers, it was not really possible to offer services to those customers based on a system built on different structure.63 Finally, in Microsoft the General Court (then the Court of First Instance) also found that a refusal to license so-called application interface protocols (API) constituted an abuse under Article 102 TFEU.64 The APIs were also found to be indispensable as it was not possible to otherwise make programs (network server operating system) interoperable with Microsoft operating system (Windows). As Windows was found in approximately 90% of personal computers (PCs), a competitor on the network server market could not compete effectively without making their products with Windows. The APIs were probably covered by both copyright and patents. Arguably, the situation in Microsoft was somewhat different in comparison with the situations in Magill and IMS Health. In theory, there were alternative ways of achieving interoperability and it could also be discussed whether interoperability with Windows was necessary for companies to be active in the network server operating systems market.
What follows from cases on IPRs and indispensability is that the threshold may not be perceived as so high as when compared to other refusal to supply cases. The very purpose of an IPR is to grant a legal monopoly that permits its holder to prohibit the use of the protected subject-matter. If that subject matter is a necessary input, and it is not possible to compete on the market without such input, the indispensability requirement is met. Arguably, the situation may be somewhat different between different IPRs. It is unlikely that access to a trademark is deemed as indispensable. By contrast, a copyrighted artistic work which has a technical function may easier meet the requirement of indispensability.
As regards patents (which is more interesting for the purpose of this paper), it is not unusual that there may be competing technologies that are protected by different patents. In addition, the distinction between different types of patents, process or product patents, may also affect the assessment under the indispensability requirement. If a particular product is covered by a patent, which constitutes its own market, the indispensability requirement will be met. As regards process patents, the situation is different as it is likely that there may several alternative processes that may be used to manufacture a certain product. That process patents may grant less protection, is evident from antitrust cases on pay-for-delay where pharma companies have obviously attempted to protect their products with agreements as process patents have been possible to circumvent.65 In addition, as regards technology that is essential to a standard (standard essential patent — SEP), it follows that users cannot substitute the protected technology, unless there are other competing standards that could be used to enter the market.66 Accordingly, it is more likely that an SEP will be deemed as indispensable when compared with other types of patents.
Another essential requirement in refusal to license cases concerns the prevention of the marketing of a new product. Importantly, if the case law on refusal to license is compared to case on refusal to supply, this requirement is what provides or should provide IPRs with extra protection.67 It could however be discussed to what extent this requirement actually provide protection in cases concerning patent as discussed below.
The criterion of the prevention of a new product was established in Magill. As explained above, in that case the refusal to license copyright protected lists directly prevented the compiling of those lists into a weekly TV-magazine. The idea behind the requirement of the prevention of new products could be seen as an expression that in the balance between the competition and IPRs, the consumer’s interest in innovation ultimately takes precedence over the interest protected by IPRs. In Magill, in particular the judgment by the Court of First Instance, the discussion of possible objective justifications did not indicate that the interests to protect by the copyright in question would be under threat.68 Thus, Magill and the following case law should not be interpreted as that competitive harm to innovation will always triumph the interests of the IPR holder. A balance must be made. On one hand is necessary to establish that new products and services are prevented, while on the other hand it must be assessed that the IPR holder cannot objectively justify its refusal. In particular, an objective justification may be raised on the grounds that the interests protected by the IPR may be undermined. While this may sound like a fair balancing exercise, it is not certain that the balancing is so fair after the Court’s judgment in IMS Health. As explained above, the rival company that requested a license had basically used the same structure for its own database that it offered for sale. While the Court probably did not intend to depart from its approach in Magill, it could be discussed whether the Court’s statement de facto lowered the requirement of a new product. The requirement would be met as long as the requesting company’s product would not essentially duplicate the product/service supplied by the dominant undertaking.69 The statement by the Court could be interpreted as meaning that a small added functionality in the rival’s product could meet the requirement. In Microsoft, the Court of First Instance definitely lowered the bar, as it declared that the new product requirement was not an absolute requirement.70 In practice, the refusal in the case prevented the marketing of already existing products, namely the network server operating system that were present on the market until Microsoft had stopped providing the APIs. It has been suggested that Microsoft gives an expression for the view that the prevention of the sales of “old products” potentially would hinder a form of follow-innovation by preventing the development of products competing with Microsoft’s own network server operating system. Arguably, the judgment seems to give a very broad view of innovation. Reading IMS Health and Microsoft, it could be debated how strong protection the case law on refusal to license actually grants to IPRs. This may be problematic, in particular, when it comes down to the protection of patent rights. Most innovation builds upon previous technology and incremental follow-on innovation may simply encompass relatively small changes to inventions covered by previous patents.
Discussing the case law on refusal to license in relation to situation in human genome editing technology two observations should be made. Firstly, as concerns the requirement of indispensability, it should be noted that although the key technology related to the CRISPR-Cas9 system has been described as revolutionizing, there are previous systems that have been used for genome editing. Zinc-finger nucleases (ZFN) and transcription activator-like effector nucleases (TALEN) are two systems that may be used for genome editing.71 From a competition law perspective, the issue arises whether the potential refusal to license could amount to an abuse in the presence of less efficient substitutes. Naturally, there is too little information to make a full-scale assessment here, meaning that the following discussion is somewhat speculative. To what an extent alternative technology actually constitutes a substitute in an individual case depends partly on the purpose the technology is used for. The CRISPR-Cas9 has been described as being potentially useful for a number of different therapeutic uses as well as use in plant technology. Arguably, the use of the CRISPR-Cas9 can be divided up in a number of different field-of-use which according to competition law probably constitute different and separate product markets. Competition in each one of those markets defined by a field-of-use is dependent on the access to genome editing technology in each respective market. Accordingly, it should be possible to capture a refusal to license only within one field-of-use. Thus, whether previous technology like ZFN and TALEN constitute substitutable technology in the sense of the Court’s statement in Bronner, depends on the specific quality of those technologies for the application within the specific field-of-use. In addition, the facts that CRISPR-Cas9 is a more efficient technology does not necessarily mean that it is indispensable. Rather, the assessment must be made objectively by determining if an efficient competitor would be able to use the alternative technology to enter and compete on the market. Only in the case that the use of the alternative technology would be costly to the extent that efficient competitor would not be able to be active on the market, the CRISPR-Cas9 technology would be classified as indispensable within that specific field of use.
Secondly, it is also interesting to address the requirement of a new product. It should initially be noted that there has not been a ‘hard’ case on refusal to license putting the interest of protecting patent rights and innovation on the one hand, and competition through innovation on the other. While such a discussion was touched upon in Microsoft, the General Court found that the parties requesting the APIs would hardly have the capability (nor the interest) to clone Microsoft’s operating system.72 Thus, similar to Magill, it could not be established that a refusal to license would de facto threaten the incentives to create of the dominant undertaking. Arguably, the situation would be quite different in the biotech sector, in particular regarding the CRISPR-Cas9 technology. It could be argued that a party requesting a license to CRISPR-Cas9 technology could in many cases claim that the refusal to license would prevent the emergence of new product (as follows from Magill and IMS Health) or follow-on innovation (as implied by Microsoft). The problem with such a reasoning is that a compulsory license would probably also threaten the patent holder’s incentives to invent. A full protection of patent rights should encompass the social value of the patent. The social value includes the value created for society as a whole and would include the potential uses of the technology that its holder have not developed. It could be argued that the very purpose of a patent is to give the patent holder control of the development of the patented technology. If the patent holder could not realize the social value of the patent, there is a risk that the initial incentives to R&D of the protected invention are not sufficiently protected, which would decrease more generally the incentives to innovate. Naturally, it could be counterargued that even a ‘compulsory license’ under EU Competition Law would encompass royalties for the patent holder. However, the literature on compulsory licensing does not express a positive view on state actors determining royalties, as these have a tendency to undervalue the patent holder’s technology. It is also important to make a distinction between Microsoft and the situation described above. Importantly, the request for a license in that case concerned the possibility for interoperability between the requesting undertaking’s products and Microsoft’s operating system, while the situation above would explicitly concern the application or the further development of the patented invention(s) on CRISPR-Cas9 technology. Arguably, the threat to incentives to innovate should be seen as more severe.
Importantly, competition law may affect the exploitation of human genome editing technology particularly by limiting and to some extent regulating the methods that holders of key patents on human genome editing technology may apply when collaborating with other market actors. In addition, EU Competition Law may theoretically also be used, in very exceptional circumstances, to get access to such key technology which would otherwise be held by a de facto monopolist. Accordingly, competition law may mainly affect the dissemination of human genome editing technology.
It follows from the discussion above that competition law may in some instances discourage dissemination because of problems of legal certainty. There is particularly an uncertainty about if and which block exemption that may apply to agreements that are part of surrogate licensing scheme. As falling outside a block exemption creates a serious situation of uncertainty about the legality of the agreements in question, considering that the application of the exemption rule in Article 101(3) TFEU is complex and difficult, it is important for technology holders to make sure that their licenses fall within one of the block exemptions. However, the scope of the block exemption regarding research (RBER) is particularly difficult to assess in relation to agreements licensing the technology to a surrogate licensor, as well as licenses between the surrogate licensor and licensees. As these agreements do not necessarily relate to research activities that are genuinely carried out jointly, they may simply not be covered by the RBER. At the same time, as such agreements may be entered into at a relative early stage of research and development, they are not likely to be covered by the block exemption regulation that applies to technology transfer (TTBER). The uncertainty regarding surrogate licensing arrangements may potentially trigger two responses. Either technology holders may apply such licensing arrangements running the risk to be struck down by competition law at a later stage or they can adapt their licensing arrangements to make them “fit” with the current block exemption regulations. Arguably, there is nothing that indicates that technology holders could not apply other form of methods of dissemination than surrogate licensing that may satisfy their needs and that will result in a widespread dissemination of human genome editing technology. On the other hand, it may also be that surrogate licensing schemes are the most efficient way for technology holders to arrange licensing to potential users of the technology and to diversify the follow-on research and the use of the technology amongst several users that in turn are specialized in particular branch of the technology. Surrogate licensing may also permit the technology holders to keep sufficient control over certain parts of the technology, while granting access to other fields-of-use to other market actors. It would be counterproductive for competition law to discourage an efficient form of disseminating the technology and it may also complicate arrangements that would promote further follow-on innovation or the further development of the commercial applications of human genome editing technology. From a wider perspective, competition law could potentially delay or obstruct the spread of technology and thereby some of the benefits to public health that the technology is supposed to deliver.
Another type of collaboration discussed above which give rise to issues of uncertainties under EU Competition Law is the potential use patent pools for disseminating key technology. While patent pools have always been controversial in EU competition law, it cannot be denied that this form of IPR management is an effective way of opening access to may potential users, which may lead to a widespread use of the technology that can lead to more competition in particular branches of human genome editing technology as well as faster development of commercialization of applications of the technology that may reach end-consumers. The current state of the competition rules on patent pools seems particularly be adapted to the needs of the telecom and electronics industries. While patent pools do not enjoy that same legal certainty as other licensing agreements by block exemptions, the Commission has at least provided a so-called safe harbor for patent pool arrangements that fulfil certain conditions. However, it is doubtful whether those conditions can be easily met by a patent pool in the biotech sector, in particular as concerns the requirement that the pooled technology should be limited to essential patent rights. If this preliminary conclusion is correct, no safe harbor for patent pools may be available for the management of key patent rights on human genome editing technology. This may, potentially, close the door for an efficient method of disseminating the technology in question. Importantly, the importance of patent pools should not be overstated, as it is uncertain whether this type of licensing arrangements ultimately will be widespread. At the moment, there is only one patent pool on human genome editing technology, and there have been doubts expressed about its future success.73
What the discussion above demonstrates is that there is unclarity about the application of competition law to different methods of disseminating technology that seem to be important in this particular sector. While it has not been argued above that the application of competition law will necessarily strike down these forms of dissemination, the legal uncertainty is problematic as such. This may call for a calibration of the current rules in competition law, in particular the scope of the RBER. However, the revision of the RBER is currently at a late stage and the discussion so far does not seem to have been focused on the particular needs in the biotech industry. However, there is a possibility for the Commission to also make revisions in its soft law that accompany the RBER. It may be good idea for the Commission to anticipate issues in the biotech industry, like licensing arrangements related to R&D cooperation regarding human genome editing technology, which are likely to raise competition law issues in the coming ten years.
Finally, it should also be noted that competition law may open up, although in very exceptional circumstances, for the possibility demanding a compulsory license from a patent holder that refuses to license key technology that is indispensable for the production of certain novel goods or services that will satisfy demand for end-consumers. This application of competition law may hinder IPR holder to monopolize (existing or potential) markets for products and services that do not merely reproduce what the IPR holder already offers, but which provides an added value to consumers for which there is demand. Arguably, this line of case law within competition law could potentially be relevant for human genome editing technology, as the key patents for CRISPR-Cas9 technology, can be used for a wide set of potential uses, some of which have not been developed yet. Additionally, it is unlikely that patent holders of such technology could exploit on their own the technology for all possible uses. Accordingly, competition law could be used to force licensing in order to promote competition and to hinder a patent holder from becoming a gatekeeper. From a societal perspective, such an application of competition law could speed up dissemination that could ultimately benefit consumers by e.g. allowing the development of treatments or plant varieties, activities which the patent holder would either not have engaged in or monopolized. However, to grant compulsory licenses may also have a negative effect on patent holders’ incentives to engage in R&D of innovations such as the CRISPR-Cas9 technology. Arguably, there is no “hard” case under competition law where the granting of a compulsory license would truly negatively affect the incentives for developing a patent protected technology. In this author’s view, the possibility to force licensing with the help of competition law should be used restrictively to technology such as CRISPR-Cas9, as it may otherwise open up for abuse from those that may want to get access to the technology. A patent holder should as a starting point legitimately be able to hold on to the exclusivity granted under a patent right and to be able to control the further development and application of the technology. If a compulsory license is to be granted under competition law, it should be established that such a license will truly bring benefits to consumers by the introduction of something novel (for instance, by hindering a particular new treatment to a disease or the development of a new plant that is more resistant to certain diseases), when the reward for developing the patent protected technology is genuinely at risk. Such a view would probably require a stricter assessment of certain conditions for finding an abusive refusal to license and a more lenient approach as regards the possibilities to justify such a refusal under the current case law.
O. Feeney, J. Cockbain and S. Sterckx, ‘Ethics, Patents and Genome Editing: A Critical Assessment of Three Options of Technology Governance’, Frontiers in Political Science 3 (2021) 731505, p. 3.
Feeney et al., supra note 1, p. 4; L. Grobler, E. Sulemanb, D.B. Thimiri and G. Raja, ‘Patents and technology transfer in CRISPR technology’ in: V. Singh (ed.), Reprogramming the Genome: Applications of CRISPR-Cas in Non-mammalian Systems Part B (San Diego, CA: Academic Press, 2021), Chapter 7.
See, e.g., Case C-457/10 P AstraZeneca AB and AstraZeneca plc v European Commission, EU:C:2012:770.
D. Matthews, A. Brown, E. Gambini, A. McMahon, T. Minssen, A. Nordberg, J.S. Sherkow, J. Wested and E. van Zimmeren, ‘The Role of Patents and Licensing in the Governance of Human Genome Editing: A White Paper’ (2021), Queen Mary Law Research Paper no. 364/2021, pp. 14–23; D. Mathews, ‘Access to CRISPR Genome Editing Technologies: Patents, Human Rights and the Public Interest’ (2020), Queen Mary University of London, School of Law Legal Studies Research Paper No. 332/2020, pp. 12–17.
Grobler et al., supra note 2, p. 168; P. Neville, ‘MPEG LA’s Use of a Patent Pool to Solve the CRISPR Industry’s Licensing Problems’, Utah Law Review 2 (2020) 535–567.
J.S. Sherkow, ‘Patent Protection for CRISPR: an ELSI Review’, Journal of Law and the Biosciences 4 (2017) 565–576; Matthews et al., supra note 4, p. 42.
Grobler et al., supra note 2, p. 167.
See, e.g., Communication from the Commission, Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements, OJ  C 89/3 (TT Guidelines), paras 208–215.
Matthews et al., supra note 4, p. 46.
Matthews et al., supra note 4, p. 42.
An overview of the review may be found at https://ec.europa.eu/competition-policy/public-consultations/2019-hbers_en.
Commission Regulation (EU) No 316/2014 of 21 March 2014 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of technology transfer agreements Text with EEA relevance OJ  L 93/17 (TTBER).
Commission Regulation (EU) No 1217/2010 of 14 December 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to certain categories of research and development agreements OJ  L 335/36 (RBER).
Article 1(1)(a)(iii)–(iv) and (vi) RBER; Article 2 RBER.
Article 1(1)(a)(i)–(ii) and (v) RBER.
Communication from the Commission, Guidelines on the application of Article 101 of the Treaty on the Functioning of the European Union to technology transfer agreements, OJ  C 89/3 (TT Guidelines).
Communication from the Commission, Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements Text with EEA relevance OJ  C 11/1 (Horizontal Guidelines).
Recital 7 TTBER; Horizontal Guidelines, para. 70.
Article 4(1) RBER.
Article 4(3) RBER.
Article 3(2) TTBER.
Article 5(b)(iv) RBER.
Article 5(2) TTBER.
Article 4(2) RBER and Article 3(1) TTBER.
Article 5(c) RBER.
Article 4(1)(a) TTBER.
Case C-73/95 P Viho Europe BV v Commission of the European Communities (EU:C:1996:405) (Viho), paras 16–17.
Article 1(1)(m) RBER.
Article 1(1)(a)(vi) RBER.
Article 3(2)–(3) RBER.
Article 1(1)(p) RBER.
See, e.g., Commission Staff Working Document Evaluation of the Horizontal Block Exemption Regulations, 6.5.2021 SWD (2021) 103 final. See also the Inception impact assessment, available online at https://ec.europa.eu/competition-policy/public-consultations/2019-hbers_en.
Case T-51/89 Tetra Pak Rausing SA v. European Commission (EU:T:1990:41) (Tetra Pak I).
Article 1(1)(m)(iii) and Article (1)(1)(n)–(o) RBER.
Article 5(b)(iii) RBER.
Article (4)(1)(c) TTBER; TT Guidelines, para 113.
Article 4(2) TTBER.
Sherkov 2017, pp. 572–573.
Feeney et al., supra note 1, p. 4.
Matthews et al., supra note 4 pp. 11–12.
TT Guidelines, para. 208.
Case C-309/99 J.C.J. Wouters, J.W. Savelbergh and Price Waterhouse Belastingadviseurs BV v Algemene Raad van de Nederlandse Orde van Advocaten, intervener: Raad van de Balies van de Europese Gemeenschap, EU:C:2002:98; Case C-519/04 P David Meca-Medina and Igor Majcen v Commission of the European Communities, EU:C:2006:492.
Grobler et al., supra note 2, p. 168, Neville, supra note 5.
TT Guidelines, para 244.
TT Guidelines, para 247.
Regulation No 19/65/EEC of 2 March of the Council on application of Article 85 (3) of the Treaty to certain categories of agreements and concerted practices, OJ  36/533, Article 1(1)(b).
TT Guidelines, para 247.
TT Guidelines, para 261.
TT Guidelines, para 261.
TT Guidelines, para 245.
Neville, supra note 5, p. 545.
Neville, supra note 5, p. 553.
TT Guidelines, paras 261(b) and 262.
Sherkow, supra note 6, p. 571.
Joined cases C-241/91 P and C-242/91 P Radio Telefis Eireann (RTE) and Independent Television Publications Ltd (ITP) v Commission of the European Communities (EU: C:1995:98) (Magill).
Magill, paras 54–56.
Case C-7/97 Oscar Bronner GmbH & Co. KG v Mediaprint Zeitungs- und Zeitschriftenverlag GmbH & Co. KG, Mediaprint Zeitungsvertriebsgesellschaft mbH & Co. KG and Mediaprint Anzeigengesellschaft mbH & Co. KG (EU:C:1998:569) (Bronner), para 41.
Case C-418/01 IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG (EU:C:2004:257) (IMS Health), para 38.
Bronner, para 45.
Joined cases 6 and 7/73 Istituto Chemioterapico Italiano S.p.A. and Commercial Solvents Corporation v Commission of the European Communities (EU:C:1974:18) (Commercial Solvents), para 16.
Bronner, paras 44–45.
IMS Health, para 29.
Case T-201/04 Microsoft Corp. v Commission of the European Communities (EU:T:2007:289) (Microsoft).
Case C-307/18 Generics (UK) Ltd and Others v Competition and Markets Authority (EU:C:2020:52).
Case C-170/13 Huawei Technologies Co. Ltd v ZTE Corp. and ZTE Deutschland GmbH (EU:C:2015:477) (Huawei), paras 49–50.
As the requirement does not apply to ordinary refusal to supply cases. Microsoft, para 334.
Case T-76/89 Independent Television Publications Ltd v Commission of the European Communities, EU:T:1991:41, paras 57–59.
IMS Health, para. 49.
Microsoft, para. 647.
Feeney et al., supra note 1.
Microsoft, para 700.
Neville, supra note 5, p. 566.