Industry Standard Setting and Antitrust
Author: Tzu-Yuan Lin
Since the standard setting organizations (SSOs) began to permit patents to be incorporated into standards, the relationship between standards and intellectual property rights has become more intricate and controversial. Issues underlying patent rights have gradually shifted from acquisition or licensing of those rights, to controlling industry standards.
In the beginning of this paper, I will take the position that Antitrust Law and Intellectual Property Law are meant to complement each other rather than work against each other. The shifting of court opinions in the past decades will be cited to support my position.
Then, I will briefly introduce the standard setting process and the kinds of legal issues that arise from it. These will include, but not be limited to, failure to disclosure, patent hold-up and refusal to license, all of which could potentially trigger Antitrust Law enforcement. I will mainly use the Rambus case to address these issues and the attitude of the Federal Trade Committee (FTC) toward them.
In the end, I will combine the lessons learned from Rambus and the new standard setting trends of SSOs to propose a strategythat will hopefully can provide some guidance for those who want to get involved in the standard setting process.
We live in a world where technology evolves each and every single day, and where one of the main forces that lead toward such technological evolution is competition in the market. To survive in the face of rapidly-developing technology, each firm will employ every possible strategy to gain advantage over the others. This includes collaborating with others to knock out a dominant firm, merging with another firm in order to enter a new market, and obtaining temporary exclusive rights granted by the law – intellectual property rights (IPRs) – to limit the what its competitors can do.
To get temporary IPRs essentially means that a firm can increase the costs of its rivals by charging them royalties. This is the incentive provided by intellectual property law in order to promote innovation and increase competition. However, the fast development of technology and its complexity have intertwined with the intellectual property law system. When a firm is driven by its eagerness to dominate the market, there is high possibility that the firm will abuse its IPRs (“patent misuse”). Such IPR abuse can have substantial anticompetitive effects and trigger the enforcement of antitrust law. This can create a paradox in which antitrust law and intellectual property law actually work against each other’s legislative purpose.
In the following section, I will briefly discuss the relationship between antitrust law and intellectual property law and take the position that there should be a balance between these two legal systems instead of conflict.
 The Purpose of Intellectual Property Law
The nature of intellectual property rights is to grant the IPR owners the temporary rights to exclude others from using their ideas. The purpose of granting such exclusivity is to provide motivation and incentives to encourage inventors to keep innovating. In the U.S. Constitution, it expressly stated that the purpose of granting patents and copyrights is to “promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.” In Mazer v. Stein, the Supreme Court pointed out that the economic philosophy behind the granting of patents and copyrights is the belief that “[i]t is the best way to advance public welfare through the talents of authors and inventors in science and useful arts.” With exclusive rights, IPR owners can recoup their R&D costs through licensing and also aggressively engage in economic activities. Old inventions are eventually eliminated through competition in the market. The combination of IP rights and the market ideally create a circle of innovation lead society to evolve again and again.
 The Purpose of Antitrust Law
The intrinsic concept of modern antitrust law is that competition can bring about economic efficiency, so that a sound market filled with competition is generally more desirable. However, this relies on a basic principle of our economic system, i.e., that free competition will ensure an efficient allocation of resources in the absence of a market failure. What is hidden behind this concept is: 1) there is always a public interest concern involved; and 2) the market itself is not perfect and cannot maintain a competitive system all the time. As a result, in order to advance public interests, the government should intervene appropriately to restore the market after a failure and to maintain competition.
So, what should antitrust law aim to do? Should it focus on maintaining a market where numerous small firms compete? Or aim to maximize consumer welfare (to increase public interests)? In United States v. Trans-Missouri Freight Asso., the Supreme Court suggested the answer should be the former and ruled that a uniform rate schedule fixed by 18 members in the freight association was unlawful even if previously allowed at common law. Obviously, the Court indicated that competition among 18 smalls firms with 18 different price rates was more desirable. But, the Court later changed its position to aim at maximizing consumer welfare in Continental T.V. Inc. v. GTE Sylvania Inc., ruling that the vertical restraints on intrabrand competition served a valid competitive purpose to increase interbrand competition, and were limited so as to accomplish only this permissible purpose. So, even when a firm’s conduct is essentially a restraint on competition, as long as it increases consumer welfare, the Court will accept such an efficiency defense.
To sum up, the purpose of antitrust law now is to prevent firms in the market from acting anti-competitively and to increase consumer welfare.
 The Interaction of Intellectual Property Law and Antitrust Law
In intellectual property law, in order to provide inventors with sufficient incentives to create and innovate, the law grants them some temporary exclusive power over price or competition, such as allowing inventors to charge a negotiated royalty or to refuse to license so that their competitors will have to design around or invent around to avoid infringement, all of which will increase their competitors’ cost. On the other hand, antitrust law wants to advance consumer welfare by making sure that the market works well and is not unfairly dominated by a single firm. In this narrow sense, IPRs seem to limit the goal that antitrust law wants to achieve – free competition – because IPR itself is essentially a monopoly right and can be used by its owners to limit rivals’ ability to compete in the market. For example, if a firm controls an essential patent in a widget, it will be able to charge any firm who wants to produce the widget a supracompetitive royalty in exchange for licensing, or it can refuse to license to anyone and limit the competition for widgets in the market. As a result, we can see that while the goal of intellectual property law is to “promote the progress of science and useful arts,” it also has negative effects that run counter to free market competition. So, the key argument about the economic efficiency of IPRs will lie in balancing the social benefits from the innovation and the cost of limitation on competition imposed by IPR owners. Indeed, such balance concern is also reflected in the legislation of Congress, and can explain why IPRs are limited in scope and duration, e.g. the patent expires 20 years after the patent application was filed.
The traditional (while oversimplified) theory for the conflict between intellectual property law and antitrust law was that: IPRs are monopoly rights, and the purpose of antitrust law is to avoid monopoly. So, these two laws conflict with each other. However, the modern U.S. courts and scholars have denied such theory of conflict, and instead adopted the theory that intellectual property law and antitrust law share a common goal. Some scholars have argued that IPRs themselves are not ipso facto monopoly rights because there is a huge difference between an exclusive right in the sense of intellectual property law and the economic monopoly concerned by the antitrust law. For example, all famous singers’ CD albums are copyrighted, yet no one will seriously challenge that Michael Jackson by his album “Thriller” held a monopoly in a relevant product market, even though the Guinness Book of World Records lists “Thriller” as having sold 65 million copies as of 2007.
On the other hand, to say antitrust law prohibits monopoly is an oversimplified misunderstanding. Although the purpose of antitrust law is to promote competition, the law has never stated that monopoly itself is illegal. Instead, antitrust law tends to focus on those anticompetitive conducts designed to dominate the market, which as a result, will decrease consumer welfare. That is to say, anticompetitive effects and decrease of consumer welfare are what the law cares about, not monopoly itself. Judge Hand in United States v. Aluminum Co. of America expressed the same concern that if a single firm survived from a group of competitors by its superior skill in the industry and business acumen, then the law does not tend to condemn the result of the force that it seeks to foster. “The successful competitor, having been urged to compete, must not be turned upon when he wins.”
In summary, the purpose of intellectual property law is to grant temporary exclusive rights to encourage innovation through competition while antitrust law aims at elimination of anticompetitive conducts and increasing consumer welfare. Given that any firm’s ultimate goal is to use their products to attract consumers’ attention, a common goal can be inferred behind these two laws - to earn the ultimate wealth by producing what consumers want with high quality but at a low price. Although IPRs grant the inventors temporary exclusive rights that limit competition, it provides inventors sufficient incentives to undertake efficient production to maximize consumer welfare in the long run. So, we can say the temporary exclusivity for IPRs is a trade-off in the middle of the way to achieve its goal. Based on this view, we can conclude that intellectual property law and antitrust law should be two complementary systems. The question here is simple: “just ask yourself whether you’d rather pay monopoly price for an iPod or a competitive price for an eight-track tape player.” Most of us will choose the former without being aware that this is because we all know that through the grant of temporary monopoly power, it will create more competition and bring us more innovative products. In this sense, both intellectual property law and antitrust law serve the same goal - to encourage innovation and competition, and to eventually maximize consumer welfare.
It is settled that the market itself is imperfect, so when market failure occurs, the government must appropriately intervene in order to restore the market disorder and maintain competition. In intellectual property law, when IPR owners misuse their rights, it results in anti-competitive effects. The Federal Circuit stated in Atari Games Corp. v. Nintendo of America Inc., “[a] patent owner may not take the property right granted by a patent and use it to extend his power in the marketplace improperly, i.e. beyond the limits of what Congress intended to give in the patent laws.” When a patent owner does so, the government must intervene to limit the exercise of IPRs.
In recent decades, antitrust enforcement for IPR misuse has shifted from a rigid per se rule toward a more flexible rule of reason test. For example, the U.S. Supreme Court used to apply the per se rule to tying arrangements (the tying of patented goods to unpatented goods) because they thought patent owners could get a certain level of market power by tying. In Carbice Corporation of America v. American Patents Development Corp., the Court stated that “[t]he attempt to limit the licensee to the use of unpatented materials purchased from the licensor is comparable to the attempt of a patentee to fix the price at which the patented article may be resold.” Then, the Court went further and extended per se rule to mandatory package licensing in United States v. Paramount Pictures, Inc., ruling that, “[l]icensing feature films only in ‘blocks’ was illegal per se because it unfairly bolstered the monopoly power granted by the Copyright Act” and did not serve any consumer welfare purpose. Correspondingly, the Department of Justice (DOJ) Antitrust Division deemed all tying and package arrangements related to IPR licensing illegal per se without considering the economic effects of the agreement in the market. Here, it is clear that neither the Supreme Court nor the Antitrust Division took the potential economic efficiency of such arrangements into account.
More recently, however, the Antitrust Division has adopted an approach that is cognizant of the economic effects of these practices. In 1995, the Department of Justice and the Federal Trade Commission (FTC) issued the Antitrust Guidelines for the Licensing of Intellectual Property (Antitrust-IP guidelines), which recognized that, “although tying arrangements may result in anticompetitive effects, such arrangements can ... result in significant efficiencies and procompetitive benefits.” Similarly, a package arrangement can also benefit the licensee by reducing overall transaction costs and eliminating licensing hold-up problems, e.g., from patent hold-up. The Federal Circuit in U.S. Philips Corp. v. ITC also acknowledged such an economic efficiency defense, ruling that “[i]n light of the efficiencies of package patent licensing…, we reject the Commission's conclusion that Philips's conduct shows a ‘lack of any redeeming virtue’ and should be ‘conclusively presumed to be unreasonable and therefore illegal without elaborate inquiry as to the precise harm they have caused or the business excuse for their use. We therefore hold that …to apply the rule of per se illegality to Philips's package licensing agreements was legally flawed.’”
After realizing that they could no longer ignore the importance of economic efficiency brought by IPR-related activities, DOJ and FTC co-released the Antitrust Enforcement and Intellectual Property Rights: Promoting Innovation and Competition (Antitrust-IP report). In this report, DOJ and FTC stated that IPR-related activities, such as tying, cross-licensing, patent pools and exclusive dealing, could potentially increase consumer welfare, and therefore, are not illegal per se. Instead, antitrust enforcement agencies should use the rule of reason standard to balance procompetitive and anticompetitive effects. And when the former outweighs the later, that activity shall be justified.
 The Emerging Issue: Standard Setting
In the Antitrust-IP report, a whole chapter is used to express concern of the DOJ and the FTC about a new type of IPR-related activity - Standard Setting. Chapter 2 of the report explains:
“Industry standards are widely acknowledged to be one of the engines of the modern economy. Standards can make products less costly for firms to produce and more valuable to consumers. They can increase innovation, efficiency, and consumer choice; foster public health and safety; and serve as a "fundamental building block for international trade." Standards make networks, such as the Internet and telecommunications, more valuable to consumers by allowing products to interoperate.
“Businesses can collaborate to establish industry standards by working through standard-setting organizations ("SSOs"). During the standard-setting process, SSO members often jointly evaluate and choose between substitute technologies. This process can raise antitrust concerns, and indeed, some collaborative standard-setting activities have been challenged under the antitrust laws. Unique antitrust issues arise when the standards adopted involve, as they frequently do, intellectual property rights.”
Since an industry standard plays such an important role in modern economies, I will try to briefly introduce what a standard is, and the advantages and disadvantages that the standard setting process can bring. Next, I will focus on the antitrust issues that can arise from it and discuss how antitrust enforcement agencies and the courts deal with them.
 What is a Standard?
Interoperability (or compatibility) has always been a big concern in industry. It is settled that the best way to solve this difficulty is to establish an industry standard. Generally, a standard adopted for certain technology means a more efficient and fast development process because all firms will not have to spend extra time and money deciding which technology specification they should follow to do the relevant R&D. In addition, a standard has network effects that can facilitate interoperability among products produced by different firms, increasing the chances of market acceptance, making the products more valuable to consumers and stimulating output. For example, the standardization of U.S. railway track in 1900 is the main reason there is a thriving railroad network in U.S. today.
In the 21st century, due to the rapid development and broad application of telecommunication and software technology, society has become a digitalized economic entity in which everything is interconnected. A standard has high value of network effects and is becoming a more prevalent practice in this digitalized marketplace. For example, standards for wireless data transmission, telecommunication, and personal computer technology are all becoming indispensable for a robust and interconnected digital economy.
So, what is the definition of a standard? According to the International Standards Organization (ISO), a standard is, "a document established by consensus and approved by a recognized body, that provides, for common and repeated use, rules, guidelines or characteristics for activities or their results, aimed at the achievement of the optimum degree of order in a given context.” The European Telecommunications Standards Institute (ETSI) defines a standard as, “A technical specification approved by a recognized standardization body for repeated or continuous application, with which compliance is not compulsory….” The FTC, on the other hand, adopted a narrower concept: "Standards…establish a common mode of interaction, which enables users to understand each other’s communication." From these definitions, we can conclude that a standard is established for common and repeated use that is related to interaction, and that its value will rise accordingly when more users adopt it, just as the internet telephone system Skype becomes more valuable as new subscribers join because more existing users can be reached. Because interaction is an intrinsic feature of a standard, standard setting will be particularly important in the industries related to communication, such as the personal computer and telecommunication industries.
 Classification of Standards
 De Jure Standards and De Facto Standards
One general classification of standards is based on how a technology is adopted as a standard, including de jure and de facto standards. De jure standards are published standards that have been ratified by the ISO and/or a range of national standards-setting bodies. Such standards are established out of concern for public interests and must be authorized by the government. In addition, they are mandatory standards, which all manufacturers in the industry must comply with. In the U.S., for example, the standards related to interstate and international communications by radio, television, wire, satellite and cable are de jure standards established and authorized by the Federal Communication Commission. In Europe, the 3G-WCDMA cell phone system, as known as the Universal Mobile Telecommunication System (UMTS), is a de jure standard collaboration by the government of the European Union and leading companies in the industry. Similarly, in May 2003, the Standardization Administration of China (SAC) approved WLAN Authentication and Privacy Infrastructure (WAPI) as a national standard for all wireless devices, and decreed that by the end of 2003, all wireless devices sold in or imported into China much comply with the WAPI standard, saying that instead of the 802.11 wireless LAN standard of Institute of Electrical and Electronics Engineers (IEEE ), WAPI is much more preferable for the sack of national security concern.
On the other hand, de facto standards are widely used voluntary standards, generally published but not yet ratified by ISO or a national standards setting. A de facto standard is usually established through the operation of market mechanism, e.g by the business acumen of a company or a company’s dominant position in the market. In 1970's, for example, there were two video tape standards competing in the market: one was VHS developed by Victor Company of Japan (JVC), and the other was Sony’s Betamax. By the 1990's, VHS had become a de facto standard format for consumer recording and viewing purposes after competing in a fierce standard war with Sony's Betamax. Another example is Microsoft’s Windows Operation System and Intel’s Central Unit Processor technology (CPU), the so-called “WINTEL” de facto standard, which was establish by these two companies’ dominant market power and advance technologies.
 Economic Efficiency of Standard Setting
Standard setting may decrease the choices available to consumers. The best examples of this are Microsoft Windows Operation System and Office Software. In fact, there are good alternative products out there in the market, such as the Apple Macintosh Operating System. In addition, a winner of the "standards war" could also manipulate its newly-acquired market power to prevent others from competing. On the other hand, standards are often procompetitive because they are designed to curb problems associated with network markets and interoperability requirements. Standards can also facilitate competition among competitors who are vying to have their technology selected as the "winning" standard. However, just like what the Antitrust-IP report mentioned, when standards incorporate technologies that are protected by IPRs, a potential "hold up" problem caused by IPR owners may arise. In the following, I will brief the efficiencies brought by standard setting, and focus on the “hold-up” problems.
 Efficiencies for Consumers
A standard can decrease transaction costs for consumers and maximize the usefulness of the products they purchase. For example, in the VHS/Betamax scenario, at the beginning of the standard war, the market was split into two, in which around 50% market share was taken by each. Under such circumstances, unless a consumer could afford and was willing to buy both products, he would have to spend extra time and money searching for the information about which video system was the best for him. In addition, to purchase either one meant the consumer would lose the benefits provided by the other, i.e. a consumer who purchased Betamax would lose the chance to enjoy the films made for VHS. And history repeats itself. Exactly the same scenario happened 4 years ago between HD-DVD (led by TOSHIBA) and Blu-ray DVD (led by SONY), and this time SONY learned from its previous experiences, and won the standard war.
 Efficiencies for Manufacturers
A standard can help manufacturers to achieve economies of scale and cost efficiency. For example, under the Blu-ray/HD-DVD scenario, unless a manufacturer had sufficient capital to license from both campaigns, it could only choose to license from one and lost the opportunity to make profits out of the other. Therefore, manufacturers would not be able to achieve enough economies of scale to get cost efficiency, and therefore, would be unable to decrease its average cost to a desirable level.
 The Dilemma Between an SSO and Its Participants During the Standard Setting Process
Before establishing a standard, the participants of an SSO will gather to discuss the specification of the standard and run through a series of procedures. Take the International Standard Organization (ISO), for example. When the ISO is establishing a new international standard, it has to run through a six-step process:
1. Proposal stage: to confirm that a particular International Standard is needed.
2. Preparatory stage: to prepare the best technical solution to the problem being addressed.
3. Committee stage: to distribute the proposal standard for the comment of the committee and, if required, voting.
4. Enquiry stage: to distribute the proposal standard for the comment of all ISO members and, if required, voting.
5. Approval stage: to distribute the proposal for the final yes/no vote.
6. Publication: to send the final approved proposal to the ISO's Central Secretariat which publishes the International Standard.
This six-step process is a model procedure and is followed by most SSOs. After the proposal stage, each SSO establishes its own rule regulating what kinds of related technologies should be disclosed before they choose a standard. A problem can arise here: because such related technologies are very important for a firm to establish its IPR strategy to compete in the market, a firm will only be willing to reveal limited information to the SSO. Therefore, how much information a firm should disclose becomes a big concern in the standard setting process. In the following sections, I will take different points of view to explain why this is a big concern both for the SSO and its participants.
 The Point of View from an SSO
In general, an SSO prefers a broader disclosure policy because this helps it to gather all necessary information and decide what are the most appropriate technologies that should be incorporated into a standard. Broader disclosure also avoids potential hold-up problems after the standard is established. As a result, an SSO wants the disclosure to be as complete as possible, including every possible technology related to the standard in progress, such as issued patents, pending patent applications, patent continuations, or even developing technology.
However, such a broad disclosure policy also has disadvantages because it may put the standard setting participants at great risk of revealing their valuable know-how to their competitors. With too broad a disclosure requirement, no firm would have an incentive to participate in standard setting. Instead, they would try to get their technology patented or copyrighted and create their own standard. Take the digital memory card market in 2000, for example, when Matsushita, SanDisk and Toshiba Corporation announced to formation of the “SD Association” in order to promote their Secure Digital Card (SD card) standard. SONY chose not to participant in it, and instead established the Memory Stick (MS) standard to compete with the SD card.
 The Points of View of Participants
Generally, participants prefer a narrower disclosure policy because of uncertainty about whether their technologies (whether protected by IPRs or not) will be incorporated into a standard. Under such circumstances, a narrower disclosure policy can at least ensure that a firm does not have to join the standard setting process at the risk of unnecessarily revealing valuable know-how to their competitors. A participant must consider how much information to disclose without putting itself in a disadvantaged position while simultaneously accessing pioneer technologies through the standard setting process. For example, during the standard setting process of computer Digital Encryption Standard (DES) in late 2000, one of the leading participants, HITACHI Corporation, owned an array of patented technologies to DES standard and attempted aggressively to convince the National Institute of Standards and Technology (NIST; also a national SSO), to adopt the same. As a result, HITACHI Corporation broadly disclosed its technologies that were relevant to DES standard. Unfortunately, NIST eventually announced its selection of a new standard that would replace the prior DES standard. This was mainly because most SSOs would like to see that the technologies they have adopted do not infringe any existing patents, so that they could avoid the licensing and royalty rate issues in the post- standard setting stage. HITACHI suffered a great loss because all of its investment in DES standard became worthless and it could not recoup all of the sunk cost it had already invested.
On the other hand, a narrower disclosure policy will lead to potential hold-up problems. After the standard is established, the manufacturers in the industry will begin to invest huge sunk costs to produce products that comply with the standard, such as purchasing fixed assets, hiring employees and making marketing expenditures. At that moment, the switching cost for the manufacturers becomes extremely high, and they find themselves “locked in” to the standard. If some participants who own essential IPRs for practicing the standard did not disclose their rights during the standard setting process, they will be entitled to sue anyone who wants to practice the standard, and may be able to hold-up the standard by charging supracompetitive royalties. Such exclusionary conduct is called “hold-up” and under such circumstances, the precompetitive effects of standard setting cannot be realized.
 Summary of Both Sides
Nowadays, both SSOs and their participants have not decided whether to go for a broader disclosure policy or not. If we recognize that the ultimate purpose of standard setting is all about cost-efficiency and economies of scale, a broader disclosure policy seems more desirable. On one hand, a broader policy can help SSOs to avoid hold-up problems. On the other hand, a participating firm will have access to the pioneer technologies of other participants. The reason many large firms join the standard setting process is because they want to make sure they can maintain a leading position in the industry. The potential benefits from being a leader in the industry can outweigh the potential business risk from disclosure. As a result, although a broader disclosure during the standard setting process may be somehow risky to a participant, it is still the worth trade-off for them to make.
 Threatening from Essential Patent Holders: Hold-Up
After a standard has been adopted, the switching cost to an alternative standard is extremely high, and the holder of IPRs that cover technology needed to practice the standard can force existing users of the technology to choose between two unpleasant options: "You either don't make the standard or you accede to it.” Under such circumstances, an owner of IPRs incorporated into a standard may have incentives to abuse its position, such as charging unreasonable royalties or excluding others from competing, all depending on how much greater the cost of switching to an alternative standard is.
In order to prevent hold-up problems from happening, SSOs have established some regulating rules that basically include disclosure rules and licensing rules. Disclosure rules require SSO participants to disclose IPRs related to a standard under consideration. Some of them are mandatory while others are voluntary. Licensing rules limit the terms that IPR owners can demand after the standard is established. These rules can govern ex ante licensing or ex post licensing. The most common licensing term is “Reasonable and Non-Discriminatory” (RAND), or “Fair, Reasonable and Non-Discriminatory” (FRAND). Some SSOs require that the incorporated IPRs be licensed on a royalty-free basis. All of these SSO rules could raise two types of antitrust concerns.
First, the ex ante negotiation of licensing terms among participants is essentially joint conduct, and therefore could constitute a violation of section 1 of the Sherman Act. However, the DOJ and FTC expressed their positive attitude toward such conduct in the Antitrust-IP report, saying that joint ex ante negotiation of licensing terms by SSO participants can be procompetitive and is unlikely to constitute a per se antitrust violation. The DOJ and FTC will usually apply the rule of reason when evaluating joint activities that avoid hold-up problems by allowing potential licensees to negotiate licensing terms with IPR owners in order to practice the standard. Such ex ante negotiations of licensing terms are most likely to be reasonable when the adoption of a standard will create or enhance market power for IPR owners. In addition, if an IPR owner unilaterally announced licensing terms, he will not violate section 1 of the Sherman Act. Similarly, if he unilaterally announced price terms, without more, he will not violate section 2 of the Sherman Act, either. Finally, bilateral ex ante negotiations about licensing terms between a SSO participant and an IPR owner outside the frame of SSO are unlikely, without more, to incur any special antitrust scrutiny because IPR owners are merely negotiating individual terms with individual buyers.
The second antitrust concern is related to violations of Section 5 of the FTC Act. These violations may arise when an SSO participant intentionally or unintentionally under-disclosed or engaged in fraud regarding the existence of its IPRs during the standard-setting process, then later alleged that any practice of the standard infringed its IPRs and required the payment of a royalty for a license. These happen because SSO rules are usually ambiguous, and because SSOs themselves do not have a powerful enforcement mechanism. As a result, this leaves some room for certain speculative participants to “play around” the rules. So far, the FTC has prosecuted three cases based on violations of Section 5 of the FTC Act, and I will introduce them in the following section.
 Cases in History
 Dell Computer (1996)
The first case about standard setting that was prosecuted by the FTC under Section 5 of the FTC Act was in re Dell. The FTC found that Dell participated in the Video Electronics Standards Association (VESA) and when asked twice, fraudulently concealed its pending patent application which, when issued, covered the established standard. After VESA adopted the standard, Dell allegedly demanded royalties from those participants who used its technology to practice that standard. Although Dell reached a consent agreement with FTC not to enforce its patent for 10 years, the case left some open questions:
(1) The representatives from Dell were engineers in the R&D department, so they were unlikely to fully understand Dell’s patent strategy and the status of its patent applications. Hence, they were not competent to answer the inquiries from VESA, and as a result, the denials were not sufficient to constitute a fraud.
(2) The disclosure rules of VESA were ambiguous. For example, “relevant to” practice standard was very indefinite. Does a patent which can poorly practice the standard count as “relevant”? Or must a patent be substantially sufficient to practice the standard?
(3) Were those engineer representatives authorized to promise and confirm the legal obligation to VESA, and if so, did such authority constitute equitable estoppel? These issues were all left unexamined.
Unocal involved proposed low-emissions gasoline standards in state regulatory proceedings in California. During the state regulatory proceedings, Unocal presented its research results (a formula) as a non-proprietary technology, and the state regulating board used these results to establish its standards. The board itself did not established any disclosure rules. While the board was setting the standard, Unocal secretly filed patent applications to cover both its refining process and formula. In 1995, oil companies sued Unocal, seeking declaratory judgment of invalidity for Unocal’s patent. Unocal counterclaimed for willful patent infringement. Oil companies claimed Unocal should be estopped from asserting patents because Unocal failed to disclose that, while participating in the standard setting, it secretly patented its refining process and formula (or did not even invent both of them) and amended pending patent applications in order to fully cover California standards. However, the district court turned the declaratory judgment action into an infringement case, and ruled that whether there was literal infringement was a matter of fact for the jury to decide. The jury found for Unocal and assessed a reasonable royalty rate the plaintiffs should pay to Unocal. The oil company then moved for Judgment as a Matter of Law (JMOL) to overturn the jury’s verdict, but was denied by the court. In a separate proceeding, oil companies argued that because of Unocal’s inequitable conduct during the standard setting process, its patents should be held unenforceable. But the district court held that these oil companies did not meet their burden of showing inequitable conduct by clear and convincing evidence. Later on, the oil companies appealed the denial of JMOL and the inequitable conduct argument by the district court, but both were denied. Finally, the Attorneys' generals from 34 states and Washington, D.C. joined in the oil companies' petition for certiorari. The Supreme Court asked the DOJ whether it wished to take position on the petition, but the DOJ declined to do so. As a result, the petition was denied.
The result enabled Unocal to charge substantial royalties and cost consumers hundreds of millions of dollars per year. Therefore, the FTC decided to file an administrative complaint against Unocal in 2003. The complaint was dismissed in an initial Administrative Law Judge (ALJ) decision based on Noerr-Pennington and jurisdictional grounds, but the FTC subsequently reversed and remanded this ALJ decision, holding that "as a matter of law misrepresentation may sometimes vitiate the Noerr-Pennington doctrine." Eventually, the full commission held that Unocal's alleged misrepresentation to the state regulatory board was not protected as a matter of law by the Noerr-Pennington doctrine because there were ample policy grounds to support that position, and the Commission had jurisdiction over whether Unocal's actions caused anticompetitive effects. Eventually, just like the ending of in re Dell, the FTC and Unocal reached a consent agreement, allowing Chevron Corporation to acquire Unocal, provided that Chevron agreed not to enforce certain Unocal patents that potentially could have increased gasoline prices in California.
The scenario in Rambus was very similar to Unocal, but involved more legal issues that crossed borders. In 1992, Rambus participated in the Joint Electron Device Engineering Council (JEDEC) to co-establish the industry standard for computer SDRAM. Under the voluntary disclosure policy of JEDEC, participants were encouraged to disclose any IPRs that were relevant to practice the standard as much and as early as they could. During the standard setting meetings, Rambus intentionally concealed all its patents and patent applications that could cover or be amended to cover the SDRAM standard. After JEDEC found Rambus’ conduct of concealment, Rambus withdrew JEDEC in 1996. The first SDRAM standard was published by JEDEC in 1993, and after Rambus withdrew JEDEC, the advanced DDR-SDRAM standard was also published. Firms in the market began to follow these two standards to manufacture SDRAM and DDR-SDRAM. In 1999, these two new standards were widely adopted in the market, and Rambus started to threaten to sue those firms who used Rambus’s patented technology to produce SDRAM or DDR-SDRAM. This resulted in two undesirable choices for firms: they could either ask for a license or litigate. Many companies who practiced the SDRAM or DDR-SDRAM standard chose to litigate in the beginning, such as Infineon, Samsung, Hynix, and Micro. Infineon was the one who made the most diligent effort to defend against Rambus.
In the trial court, Infineon tried to use fraud as a defense, and the court found that Rambus was liable for committing actual and constructive fraud in its conduct at JEDEC, with respect to the SDRAM standard. For the DDR SDRAM standard, because it was established after Rambus withdrew JEDEC, the court found there was no fraud. However, the appellate court found that Rambus did not have any conduct of fraud and reversed and remanded the decision of the trial court.
The appellate court first defined the scope of disclosure, ruling that “Rambus 's duty to disclose extended only to claims in patents or applications that reasonably might be necessary to practice the standard.” That is, there must have been some reasonable expectation that a license would have been needed to practice the standard, and an equivalent analysis was unnecessary. After defining the disclosure scope, the appellate court went on to decide whether Rambus misrepresented during the standard setting process and ruled that:
“To prove fraud in Virginia, a party must show by clear and convincing evidence: 1. a false representation (or omission in the face of a duty to disclose), 2. of a material fact, 3. made intentionally and knowingly, 4. with the intent to mislead, 5. with reasonable reliance by the misled party, and 6. resulting in damages to the misled party. A party's silence or withholding of information does not constitute fraud in the absence of a duty to disclose that information. Generally, “fraud must relate to a present or a pre-existing fact, and cannot ordinarily be predicated on unfulfilled promises or statements as to future events.” In some cases, however, misrepresentations about a party's present intentions also may give rise to fraud. Failure to prove even one of the elements of fraud - such as existence of a duty to disclose - defeats a fraud claim. [Emphasis added]
The appellate court found that Infineon did not prove with clear and convincing evidence that Rambus’s patents or patent applications actually covered JEDEC’s two standards; in addition, the appellate noted that the disclosure policy of JEDEC was overly broad and ambiguous. Under such circumstances, the participants of JEDEC would not be able to know exactly how to follow the rules. As a result, it was difficult for JEDEC’s members to meet the clear and convincing evidence requirement.
In 2002, while Rambus and Infineon were still at trial, the FTC filed a complaint accusing Rambus of violating Section 5 of the FTC Act. According to the FTC, Rambus acquired monopoly power through fraud and exclusionary conducts during the standard setting process in JEDEC. In addition, because firms had invested huge sunk costs in order to practice the standard, their switching cost became extremely high. Such lock-in effects made Rambus’s monopoly power even more durable. As a result, the FTC concluded that Rambus unlawfully monopolized the markets for patented technologies incorporated into JEDEC’s standards in violation of Section 5 of the FTC Act. The initial ALJ decision was released in February 2004, in which Judge Stephen McGuire ruled that the Complaint Counsel had failed to sustain their burden of proof with respect all violations in the Complaint. The result and reasoning were similar to the decision of the appellate court in Rambus, Inc. v. Infineon Techs. AG. Both the ALJ and the appellate court found that Rambus did not intentionally conceal relevant patents or patent applications while participating in JEDEC. As a result, there were no grounds on which JEDEC and its participants could claim any damage from Rambus’s misrepresentation.
However, after the appellate court remanded the case to the district court, Infineon took another strategy to compel the production of various documents Rambus was withholding on the basis of the attorney-client and work product privileges. The district court ordered that Rambus produce to Infineon any documents it produced in the Hynix or FTC litigations. Rambus was also ordered to produce 27 documents that Rambus had never disclosed previously. The collection of these newly-revealed documents enabled Infineon to raise a strong unclean hands defense because the documents demonstrated that Rambus engaged in unlawful spoliation of evidence about its patent applications and communications with its patent attorneys. Later on, a bench trial was held on Infineon's defense of unclean hands and with respect to Rambus’s spoliation of evidence. The Court ruled “[f]rom the bench that Infineon had proven, by clear and convincing evidence, that Rambus was liable for unclean hands, thus barring Rambus from enforcing the four patents-in-suit. Additionally, the Court ruled that Infineon had proven, by clear and convincing evidence, that Rambus had spoliated evidence…” Following this ruling, Rambus and Infineon settled their litigation before the Court issued its written opinion. In Samsung, the Court also held the same.
However, there were at least two courts that held that the conduct of Rambus did not constitute unlawful spoliation of evidence, such as in Hynix and Micron. Because Rambus series cases had a huge impact on the SDRAM and DDR-SDRAM industry, the FTC decided to take a further step to appeal the initial ALJ decision. In the appeal process, in order to avoid the previous discrepancy of the courts’ opinions about spoliation of evidence, the FTC focused on how to establish the casual link between Rambus’s exclusionary conduct and its acquiring monopoly power under Section 5 of the FTC Act. In deciding whether there was exclusionary conduct, the FTC used the following analyses:
1. Relationship between Patent and Antitrust Law in Cases Involving Fraud on the Patent Office or Patent Enforcement Initiated in Bad Faith.
2. Standard of Proof Should Be Commensurate With Proposed Remedy.
3. Chilling Participation in SSOs.
4. Reliance on Testimony Rather than Contemporaneous Written Evidence.
In applying the analyses above, the FTC defined exclusionary conduct as, “conduct other than competition on the merits – or other than restraints reasonably ‘necessary’ to competition on the merits – that reasonably appear[s] capable of making a significant contribution to creating or maintaining monopoly power.” The alleged exclusionary elements here were that Rambus engaged in a course of deceptive conduct in order to influence JEDEC to adopt certain standards. For conduct to be found deceptive, “there must have been a ‘misrepresentation, omission or practice’ that was ‘material’ in that it was likely to mislead ‘others acting reasonably under the circumstances and thereby likely to affect their conduct or decision[s].’” Based on the finding that, when asked at two JEDEC meetings, Rambus’s intentional concealment misled JEDEC and its participants to believe that Rambus did not have any related patents or patent applications that covered the drafting standard, the FTC concluded that there was causation between Rambus’s conduct and its monopoly power. Therefore, Rambus was in violation of Section 5 of the FTC Act.
Although the FTC concluded that Rambus had violated Section 5 of the FTC Act, Rambus’s related patents to practice SDRAM and DDR-SDRAM standards were still valid and enforceable under patent law. To solve this problem, the FTC ordered both sides to submit reasonable royalty rates for the Commission to decide. The Commission reviewed rates from other comparable licenses in the industry and found that “a maximum royalty rate of .5% for DDR-SDRAM, for three years from the date of the Commission’s Order and then going to be zero, is reasonable and appropriate.” It also found that “a corresponding .25% maximum rate for SDRAM is appropriate.” This eventually put an end to the issues arising from SDRAM and DDR-SDRAM standards. Some commentators suggested that these royalty rates (0.5% and 0.25%) indicated that at a certain level the FTC condemned the inequitable conduct of Rambus because in general practice, the average for a single-patent “reasonable royalty” damages case is around 13%. In the Information Technology (IT) industry, the average royalty rate is 7%. Therefore, the below-average royalty rates determined by the FTC could be deemed as punishment for Rambus.
The Rambus case was a big step toward the FTC taking an aggressive attitude in emerging hold-up issues. Under the flexible standard adopted by the Commission to establish a casual link between Rambus’s misleading conduct and its monopoly market power, it will be easier for the FTC to bring a successful prosecutions in the future under Section 5 of the FTC Act. But the fundamental problem that led to the issues in Rambus has not been resolved – i.e., the ambiguity of disclosure and licensing rules of SSOs. And as long as the ambiguity exists, similar issues will arise again in the future. In light of this, the Opinion of the Commission noted that SSOs should try to eliminate ambiguity and establish a definite disclosure duty for their participants to follow.
 Worldwide Trends
 Trends in the U.S.
 VMEbus International Trade Association (VITA)
Located in the U.S., VITA is a non-profit organization of vendors and users having a common market interest in real-time modular embedded computing systems. VITA has devoted itself to creating open standards and was recognized by the American National Standardization Institute (ANSI) and the International Electrotechnical Commission (IEC). After the Rambus case, VITA has been developing new disclosure policy and licensing terms, the most important of which is its radical “mandatory ex ante” disclosure rule. According to VITA’s patent disclosure policy, any member whose patents or applications that are believed to contain claims that may become essential to the drafting standard of VITA in existence at the time, must disclose those patents or applications within 60 days after the formation of VITA’s working group or no later than 15 days from the date of publication of the drafting standard. That member also must license with FRAND terms and declare the maximum royalty rate it may charge for all claims. In addition, if any member fails in its duty of disclosure, that member must license all concealed patent claims to the extent that is essential to the drafting standard. Finally, VITA also established an arbitration procedure to deal with issues arising from its disclosure policy.
In October 2006, the Department of Justice, Antitrust Division (the DOJ Antitrust Division) rendered a business review letter, recognizing that the new disclosure policy and licensing terms of VITA could effectively prevent standard hold-up from happening, and did not have any anticompetitive intent. After the business review letter from the DOJ Antitrust Division, ANSI also accredited VITA’s new policy in January 2008. These two decisions carried a very important message that both the DOJ and ANSI were not opposed to seeing an SSO adopt a clear and rigid patent policy to create a sound environment for standard setting. This might be a touchstone for other SSOs to follow in order to prevent standard hold-up issues in the future.
IEEE has long been adopting an ex ante disclosure policy in its standard setting process. After the Rambus case, IEEE began its patent policy reform by establishing a clearer, voluntary, ex ante disclosure and licensing policy. One month after the DOJ Antitrust Division accredited VITA’s new patent policy, IEEE also asked for the same accreditation and got a business review letter recognizing its new voluntary ex ante disclosure policy as procompetitive. The new IEEE policy has three key elements, according to its news release:
(a) It permits and encourages the optional and unilateral ex ante disclosure of royalty rates and other license terms - that is, disclosure before a patented technology is included in a standard. The disclosed terms may include, for example, the maximum royalty rate that the patent holder will seek to charge.
(b) It improves the mechanisms for making sure that a patent holder's assurance (which is irrevocable) fully and effectively binds subsequent owners of the patent by requiring the patent-holder to provide notice of the existence of the assurance.
(c) It strengthens provisions for binding the submitter's affiliates to the terms of the policy, making clear that affiliates are bound unless the submitter identifies affiliates it does not wish to bind.
Compared with VITA’s new patent policy, IEEE gave more options to its members. If the chair of an IEEE standard working group believes a patent potentially will be essential to the drafting standard, the chair may ask the patent holder to disclose its relevant patents rights and to provide a letter of assurance (LOA) about licensing terms, which includes five options:
(a) Provide no assurance.
(b) State that it does not have essential patent rights.
(c) Commit not to enforce its patent rights against members who practice the standard.
(d) Commit to license on RAND terms.
(e) Commit maximum royalty rates or the most restrictive non-price terms.
If a member chooses to commit to the maximum royalty rates or the most restrictive non-price terms, the chair of the working group will provide relevant price information to all members to reconsider the possibility of an alternative standard. Finally, IEEE itself does not provide any dispute settlement mechanism, so if there is any dispute arising in the middle of standard setting, that dispute will be left for the courts to decide.
 Trends in the European Union
 European Telecommunication Standards Institute (ETSI)
Located in Europe, ETSI is has long devoted itself to standard setting in information and communication technology, including the standards for broadcast, mobile, radio, internet industry…etc, and was recognized by European Commission (EC) as an official SSO in Europe. After the Rambus case, ETSI changed its patent policy in order to clarify the ex ante disclosure and licensing issues. According to the “ETSI Guide in Intellectual Property Rights,” members are obligated to disclose the essential IPRs of theirs or others (if any) as early as possible, and to license their rights on FRAND terms. If any essential IPR holder refuses to license its rights, the chairman of ETSI must inform the Committee Secretariat to reconsider whether to suspend the drafting standard or specification. Finally, for disputes arising from ETSI’s IPR policy during the standard setting process, members should first use a friendly manner to solve these disputes or ask ETSI for mediation.
Many SSOs are now reforming their patent polices in order to prevent standard hold-up problems from happening in the future. Some of them have adopted mandatory ex ante disclosure policy while the others believe voluntary ex ante disclosure is more appropriate. So far, it is not yet settled which one is the best for standard setting. As former Deputy Assistant Attorney General of DOJ, Gerald F. Masoudi, commented in a DOJ Antitrust Division report, “no standard-setting process is perfectly efficient. No patent system has perfect patent quality. No disclosure or licensing policy can perfectly anticipate every future need of patent holders and licensees…we trust market forces to eventually, if imperfectly, reach efficient results.” Mr. Masoudi might be right because only time can tell whether an ex ante disclosure and licensing policy is the panacea for those issues during standard setting. So far, we can only count on the market mechanism and the experiences of SSOs and antitrust enforcement agencies to fix those issues.
 Suggestions For SSO Participants
Since no SSO has a perfectly efficient patent policy to manage the standard setting process and anticipate the expectations of all participants, discussion of a comprehensive patent policy is not practical. However, a firm, at least, can still try to clarify the ambiguities of an SSO’s patent policy before it participates in the SSO. In the following section, I will try to propose some suggestions based on the issues I have been dealing within this paper.
 When the Duty of Disclosure Starts
If an SSO participant discloses relevant patents too early, it will put its business interests at risk. This is so because the disclosure will allow the SSO to have enough time to look for an alternative technology or try to invent/design around the patent. As a result, it will be hard for the participant to recoup its sunk cost. On the other hand, a late disclosure may cause a participant to violate its duty and end up in litigation.
 When the Duty of Disclosure Ends
Before joining an SSO, a firm should clarify whether its duty of disclosure is a continuous duty (even after it quits) or a temporary duty while it is a member of that SSO. A continuous duty can have a great impact on a firm’s patent strategy in the future.
 To Whom Should the Participant Disclose
After becoming a member of an SSO, the participant should make sure to whom it should disclose its relevant patents. Should it disclose only to the committee of the SSO or merely disclose to other participants who have enquiries?
 How Related A Patent Must Be
Before disclosing one’s patents, a participant should ask how related a patent must be to the extent of setting the standard. Does an equivalent patent under doctrine of equivalent count as “related”? Or only patents that can read on the drafting standard? An over-disclosure can put a participant at great business risk by revealing too much information to its competitors in the market.
 Level of Care
After becoming a member, a participant should make sure whether its level of care is only a reasonable expectation standard or a strict liability standard. This will determine how broadly a participant should conduct relevant patent research.
 Scope of Disclosure
Sometimes a drafting standard may involve not only patents but also copyrights, trade secrets or other IPRs. Therefore, a participant must know to what extent it is supposed to disclose related IPRs: otherwise it may put itself at risk of litigation for failure to disclose.
 Legal Effect of Disclosure
Before disclosing one’s IPRs, a participant must make sure what legal effects its disclosure will have. That is, will the disclosure connect to licensing with RAND, FRAND or free-licensing terms? If it is connected to RAND or FRAND licensing, what standard will be used to define “reasonable?”
In addition, does failure of disclosure indicate that the patent holder covenants to waive its right to claim patent infringement in the future? This is what actually happened in Qualcomm Inc. v. Broadcom Corp. Qualcomm in this case intentionally hid its patents that were essential to the drafting standard during the standard setting process and later on sued Broadcom for patent infringement. The federal circuit adopted the “implied waiver” theory and ruled that a firm’s participation in an SSO implicates a duty to disclose its patents. Similarly, a firm’s failure to disclose will imply its waiver to those patents that are necessary to practice the standard. Therefore, Qualcomm’s failure to disclose its patents during the standard setting process impliedly waived its patent rights to the extent of practicing that standard.
 Compensation for Disclosure
A participant should determine whether there would be any compensation for disclosure if the SSO eventually did not incorporate any of the participant’s technology into the drafting standard. This is so because when a participant discloses its patents, pending applications or any other unpatented technology, it will allow its competitors to predict its future patent strategy, and therefore put the participant in a great business hazard. In addition, if the participant discloses any new technology that is yet to be patented, it will have to be cautious due to the novelty requirement, which means it has to file patent application within one year in the U.S. and may lose on the issue of novelty in foreign countries.
 Patent Pool Concern
If the participant is a licensee, it must make sure that some important companies in the industry are also included in the patent pool. Otherwise that patent pool may turn out to be worthless. For example, if a patent pool related to the standard setting for mobile phones does not include Nokia, Motorola and SonyEricsson, then that patent pool is probably worthless. On the other hand, if the participant is a licensor, then it must determine how to calculate the royalty rate. Generally, a big firm with a lot of patent rights would prefer a patent-number basis method. And a small starting firm with only a few patents would prefer a company-number basis method.
 Is the Duty of Disclosure Transferrable?
Merger and acquisition are very common in today’s business world. Therefore, if the acquired company joined an SSO that has a duty of disclosure, the acquirer had better make sure whether that duty can be transferred while conducting due diligence.
In Rambus, the royalty rates under average determined by the FTC can be deemed as punishment for the conduct of Rambus. In Qualcomm, the court ruled that, “[e]ven if implied waiver did not apply here, Qualcomm's conduct falls within another equitable doctrine – equitable estoppel” and would be barred from enforcing its patent rights to the same extent. So, we can see that not only Rambus but also Qualcomm tended to condemn conduct that was essentially inequitable. This is probably because the courts have a policy concern to avoid potential hold-up problems and promote competition. Since standard hold-up has so much negative impact on competition and will keep arising in the future, we can reasonably expect that the courts will be likely to follow the same rationale and impose punishment on any inequitable conduct that has anticompetitive effects during standard setting process in order to achieve the ultimate purpose of both Antitrust Law and Intellectual Property Law – to promote competition.