The changing landscape of standard-essential patent licensing

The good news is that a rationalised essential patent regime will likely be good for everyone – the public, licensors and licensees. Of course, this does not mean that some people will not lose out; after all, some actors gamed the system to take advantage of the opportunities of others.

The history of standard-essential patent (SEP) licensing goes back a long way and encompasses some extremely diverse technologies. It has become a hot topic in recent years, with the development and phenomenal growth of cellular communications. The public even became aware of the issues with the advent of the smartphone ‘patent wars’. This article primarily focuses on the telecoms sector, but the principles discussed apply to any technical standard.

Technological standards are particularly likely to arise in technologies where no one manufacturer is likely to produce a complete system, and where the parts and components of the various manufacturers must interface with each other in order for the overall system to function properly.

The public benefits of technological standards can best be explained by what happens in their absence. The world’s electrical power distribution systems evolved independently, leading to a multiplicity of standards for power sockets which, even 130 years later, still compels travellers to carry adapter plugs with them when they go abroad. Without mobile telecommunications standards, each manufacturer would have to operate a closed system; meaning that if one were to buy a Samsung handset, it would work only with a Samsung base station. A telecommunications network operating in this manner would be useless due to missing coverage, considerably more expensive and less robust than the system which we have today.

Patents play a significant role in conferring private benefits for participation in a technological standard to the companies that help establish it. The most relevant patents for a technology standard are referred to as ‘essential’. These essential patents become relevant in all areas in which manufacturers have agreed to standardised interfaces or specifications in order to ensure the proper cooperation of various parts of a system. Examples can be found everywhere, not only in cellular or wireless communication.

If the proper functioning of standardised technologies requires the use of specific technical solutions, the patents protecting these solutions become essential (ie, they cannot be circumvented by other possible solutions). For a company to say that its products adhere to a technical standard is to say unequivocally that its products infringe any valid patents that are essential to that technical standard (eg, “HappyCo makes 3G handsets” means that HappyCo’s 3G handsets infringe every valid patent that is essential to the 3G standard).

The advantages to owning essential patents are obvious. Manufacturers need licences to SEPs in order to make products. If it can be shown that a patent is essential to a specific standard, then all products implementing that standard necessarily infringe the patent and there is no need for further proof. For standards such as cellular communication standards, this implies that billions of devices (and the related infrastructure) are potentially infringing products – an impressive royalty base for licensing negotiations.

SEPs potentially give their owners the power to exclude competitors from huge markets, much to the disadvantage of other market participants. Both legislators and standard-setting organisations (SSOs) have identified that risk and have taken steps to try to prohibit anti-competitive behaviour by owners of SEPs. Over the last 10 years it has become increasingly difficult in many countries to obtain injunctions based on the infringement of SEPs. SSOs have established IP policies requiring participants in standards-setting procedures to agree to license their SEPs on fair, reasonable and non-discriminatory (FRAND) conditions. Some SSOs do not officially require fairness, but request the owner to license its patents under RAND conditions.

Three topics have been discussed widely as being relevant for SEP licensing: patent hold-up, reverse patent hold-up and royalty stacking. We discuss these further below and explain how they fit into a licensing discussion.

Figure 1. Prospective manufacturers want to obtain access to a licensor’s valuable technology – the licensing protocols for standard-essential patents (SEPs) follow a fair, reasonable and non-discriminatory (FRAND) approach imposed by the standards-setting organisations

© 2014 Archimedes Intellectual Properties, Ltd

SEP licensing – the practitioner’s perspective

SEP licensing is big business. The European Telecommunication Standardisation Institute (ETSI) – which collects declarations of patents for cellular communication standards – lists over 1,400 essential patent families for second-generation (2G) communication technologies (eg, GSM, GPRS, EDGE), around 3,000 families for third-generation (3G) (eg, WCDMA) and a similar number for long-term evolution (LTE) technologies. Bear in mind that these are patent families, not just patents.

The 2G technologies introduced digitally encoded telephony conversations – a more efficient use of the spectrum assigned to wireless telephony – the beginning of text messaging and, with GPRS, the first low-bandwidth packet-based data communication system. The 3G technologies, among other things, allow for an information transfer rate of at least 200 kilobits per second. Fourth-generation (4G) technologies provide mobile broadband internet access to mobile devices of all types. LTE – a version of 4G – offers a standard for wireless communication of high-speed data for mobile phones and data terminals.

ETSI registers declarations for SEPs, but does not investigate whether the patents submitted are truly essential to the relevant standard or even whether they are valid – in fact, it will even accept declarations for SEPs for pending patent applications. Of course, ETSI does not have sufficient staff to conduct these investigations – even if it did, its findings would not necessarily be binding on any court that subsequently tried the issue. Moreover, ETSI would expose itself to allegations of anti-competitive behaviour if it excluded certain patents from its database.

For a company trying to enter the cellular communication devices market, understanding and navigating the landscape of SEP licensing is a formidable task – and one that has proven increasingly difficult.

During the early days of GSM telephony, fewer than 10 patent owners had portfolios containing valuable essential patents. The number of GSM telephony manufacturers was likewise fairly small, with many also members of the SEP owners’ club and business often being conducted in a fairly collegial manner.

Today, the situation has changed significantly. The number of companies claiming to have valuable portfolios in 3G or 4G has exploded, passing the 100 mark. The manufacturers’ side has grown similarly, but only a few hold essential patents.

The telecoms licensing situation also changed substantially when computer companies began manufacturing devices with telephony functions. The advent of the smartphone expanded the number of devices needing licences to SEPs. The smartphone era also brought the computer industry within the norms established by companies operating in the telecoms sector. This culture clash forever changed the relationship between licensees and licensors.

A further change in the market has come from the many owners of essential patents which have started to sell parts of their SEP portfolios to third-party patent aggregators and licensing entities. For example, Nokia and Ericsson have both sold portions of their portfolios to a variety of non-practising entities. This trend adds to the number of potential licensors with which a market entrant must cope.

The royalty rates for mobile telecommunication have generally been declining. The overall trend has been towards lower patent royalties, in particular in the United States. To some extent, this may coincide with a few early cases in which licensors successfully obtained greater royalties than has subsequently been the case. In addition, the first generation of mobile telecoms patents has now expired and the 2G patents are moving towards expiration. Royalty stacking, discussed below, has tended to cap individual royalty rates for many licensors. Further, many licensors wanted to make 3G succeed over other technologies and early on announced that they expected fairly low royalty rates as fair value for all related patent licences. Having selected an initially low number, the licensors cannot now make this higher.

The expiration of early 2G (GSM) patents and the looming expiration of the remaining 2G (GPRS and EDGE) patents illustrate a fundamental characteristic of any SEP licensing regime. The patents germane to any standard must typically already be on file before the standard is formally adopted. This means that every technical standard will eventually become patent free. So, one day the existing mobile communication standards will be royalty free.

For every company entering a market with many SEP owners, the number of requests for licensing negotiations and royalty-bearing licensing agreements must be overwhelming.

Figure 2. Everything goes well as long as the prospective licensees can agree on FRAND terms with the SEP licensor – given the commercial significance and expense of SEP licences, the temptation for each side to game the licensing negotiations is overwhelming

© 2014 Archimedes Intellectual Properties, Ltd

Determining fair value

As with any other patent portfolio, the value of an SEP portfolio depends on several factors, among which the size of the portfolio is only one indicator. The technical merits of individual patents and questions of validity, essentiality and regional coverage must also be considered.

It is obvious that a licence to an individual SEP is of limited value for a licensee if the same licensor owns other allegedly essential patents. Signing a licensing agreement for a limited number of these SEPs would implicitly acknowledge the value of (part of) the licensor’s portfolio, while exposing the licensee to immediate new licensing requests or litigation. Therefore, a licence to an SEP portfolio makes sense for a licensee only if it covers all SEPs relevant for its products. The challenge then is to find the value of a portfolio licence rather than that of individual patents. Court decisions typically set royalty rates only for the litigated patents and are therefore of limited value when it comes to valuing an entire SEP portfolio.

Considering the large number of patents declared essential and the size of some of the larger SEP portfolios available, it becomes a real challenge for licensors to evaluate individual licence offers. The large number of patents declared essential to cellular standards may indicate that many of these patents have fairly limited inventive value. The large number of declared SEPs points to the commercial importance of this sector and creates a compelling incentive to engage in gaming.

As mentioned, SSOs normally just register declarations of essential patents. Only a court can verify the correctness of these declarations, although it cannot typically hear a case unless a disagreement has arisen. Thus, an SEP can theoretically be licensed for many years before a court finds that it is invalid or is not being infringed (which would imply that it is not needed to practise the standard).

External evaluations of the accuracy of essentiality declarations indicate that overall, no more than 50% of all patents declared essential for any cellular telecoms standard are actually essential. Based on the large number of participants in standards-setting processes, there seems to be high probability that inventions may get patented several times by several companies. Obviously, this may lead to a high percentage of patents which would not survive an invalidity challenge unscathed.

In addition, a company seeking a licence to an SEP portfolio needs to understand that many patents might be essential only to optional parts of a standard, and that a given manufacturer may not need licences to many of these patents, depending on the configuration of its products. Understandably, patent owners usually offer no support for such evaluations.

Thus far, it has always been difficult for both licensor and licensee to evaluate whether a proposed settlement value is fair, reasonable and non-discriminatory. The parties will obviously have different positions about what constitutes a fair settlement. Due to the confidentiality of most licence agreements, it is usually extremely difficult to evaluate whether a contract discriminates against one party compared to other licensees. Also, the SSOs imposing FRAND obligations on their standardisation process participants usually do not define in detail what FRAND actually means – and it would be difficult to try to come up with such a definition without potentially violating competition laws.

Figure 3. While still making standard products, prospective licensees may evade their responsibilities to obtain SEP licences from the licensor. This creates a strong disincentive for the licensor to participate in future standard-setting activities and may signal to other companies that participating in a standard-setting activity is not commercially recommended

© 2014 Archimedes Intellectual Properties, Ltd

Specific issues in FRAND/SEP licensing

Three specific issues in FRAND/SEP licensing have been particularly subject to gaming over the years by licensors and/or licensees. Fortunately, these are the very issues that appear to be in the process of being rationalised.

‘Patent hold-up’ describes the situation whereby an SEP owner requests unreasonably high royalties for the use of its patents, knowing full well that the infringing party cannot avoid the use of the patents when manufacturing standards-compliant products or systems. This risk has been reduced by many courts, which have decided that injunctions based on infringement of SEPs might be anti-competitive and should therefore be awarded only in very specific circumstances. In most jurisdictions, an SEP owner risks various counter-defences based on anti-competitive behaviour that run independently of questions about infringement.

‘Reverse patent hold-up’ describes the behaviour of a potential licensee which tries to stall negotiations about a FRAND licence agreement with a licensor, even where it is not disputed that the patents are essential and used in the prospective licensee’s products. The fact that, for competitive reasons, the patent owner cannot exclude the infringing products from the market gives the infringing party (theoretically) the possibility of implementing the patented technology without ever paying royalties. This problem has been addressed by a number of courts and competition agencies, resulting in a number of decisions and agreements which have helped to reduce the risk of SEP owners becoming victims of reverse hold-up actions. One example of an authority laying out rules to proscribe reverse patent hold-up can be found in the European Commission’s decision in In re Samsung and the Wathelet opinion.

‘Royalty stacking’ describes two issues on which there is no legal consensus with regard to whether the problem exists in reality or only in theory; yet there are complaints from prospective licensees regarding the level of specific royalty rates. These two issues are as follows:

• The use of many technologies in complex devices, such as smartphones, may lead to an extreme royalty burden on the finished products because the manufacturer must pay royalties for all patents used in a product. Studies – which remain contested – conclude that the royalty burden on a smartphone might well exceed the manufacturing cost of such a device.
• Even if looking at just one technology area (eg, cellular communication), royalty stacking can become a potential issue. If all patent owners were to request royalties on the use of their SEPs, the burden on any product might become economically intolerable to the point that it was not possible to realise a profit from manufacturing it.

Apportionment offers a possible solution to the first issue and is becoming the norm applied by courts in many countries. For the second issue, courts in some countries have rendered decisions (while still under appeal) setting royalty rates for specific portfolios which, if applied to all essential patents, would limit the overall royalty burden to an acceptable level. A number of SEP owners in the cellular communication industry have published their view on the appropriate aggregate royalty burden for 3G and LTE. While these published values remain disputed, they offer an indication to companies seeking licences of what value might be considered fair and reasonable for a specific portfolio.

Obviously, the acceptable royalty burden on a product has upper limits. In markets with narrow profit margins, it becomes economically unfeasible to sell products if the potential profit is below the royalty burden imposed through patent licences and consumers are unwilling to pay higher prices based on these royalties. Many high-tech markets show a strong trend towards low profit margins. The smartphone market is not excluded from that trend, although it still seems fairly resistant due to the continuous integration of new features and functions.

Moreover, companies may have to compete with others which pay much smaller or even no royalties (either because they have their own patents based on their own R&D or because they operate in countries with more lenient patent enforcement systems), and can therefore offer their products at lower prices. Some surveys show that at certain points, as much as 40% of the telecoms market may not have had the required SEP licences. While probably overstating the case, few would argue that the vast majority of telecoms devices are inadequately licensed. If the royalty burden of a product exceeds the possible profit, a diligent manufacturer will naturally consider leaving that particular market. Therefore, royalty stacking may be or become an issue, particularly for market participants with lower profit margins.

Figure 4. The SEP licensor may make onerous demands of the prospective licensees – either so outrageous that no commercial deal can be struck or so high that the licensee cannot manufacture a product and make a sufficient profit. Some licensees may even be upset about having to pay anything for the rights to 20-year-old technology

© 2014 Archimedes Intellectual Properties, Ltd

More to FRAND than just royalty rates

FRAND-compliant licensing agreement must fulfil other quality criteria than just defining a mutually acceptable royalty rate or royalty value. Some SSOs allow the patent owner to make its FRAND offer subject to licence reciprocity. In that case the value and extent of the back-licence need to be considered during FRAND negotiations.

In case of cut-off licence agreements (those with a clear termination date), it should be clear that the parties will negotiate a FRAND-compliant extension of the contract. A ‘catch-up’ clause is typically found in large commercial SEP licences and provides a mechanism for the parties to discuss the new patents in their respective portfolios when the licence expires after a defined period. The option to request the renegotiation of royalty rates in case of significant changes in the portfolio value should be considered.

Many licence contracts explicitly prohibit invalidation challenges initiated or supported by the licensee. In Germany, the so-called Orange Book offer explicitly requests such a non-aggression clause. Orange Book (KZR 39/06) considered conditions under which a compulsory licence to an essential patent might be adequate.

However, in his response to the Dusseldorf questions, General Attorney Wathelet of the European Court of Justice (ECJ) clearly expressed his view that such a non-aggression clause would not comply with EU competition law. While we do not know with certainty that the ECJ will follow Wathelet’s opinion in its decisions, we may assume that any non-aggression requirement would not comply with FRAND or competition law.

Figure 5. The regulator may come to the aid of the SEP licensor when a prospective licensee refuses to pay outright or effectively refuses to pay for the SEP licences by failing to allow negotiations to complete

© 2014 Archimedes Intellectual Properties, Ltd

Dusseldorf questions

Germany has long been an interesting venue for patent infringement cases because of the comparable ease with which an injunction can be obtained if a patent is found to be valid and infringed. This situation changed in 2013, at least for SEPs. Before 2013, the only way for an infringer to avoid an injunction was to make an Orange Book offer. However, since 2013 the courts have been reluctant to enjoin infringers and have stayed many cases involving essential patents.

In March 2013 the Dusseldorf Higher Regional Court stayed a case between Huawei and ZTE (after finding that the patent in suit was infringed) before issuing an injunction against the infringer. Instead, it asked the ECJ’s opinion on the following five questions:

1. Does the proprietor of a standard-essential patent which informs a standardisation body that it is willing to grant any third party a licence on [FRAND] terms abuse its dominant market position if it brings an action for an injunction against a patent infringer even though the infringer has declared that it is willing to negotiate concerning such a licence?

or

is an abuse of the dominant market position to be presumed only where the infringer has submitted to the proprietor of a standard-essential patent an acceptable, unconditional offer to conclude a licensing agreement which the patentee cannot refuse without unfairly impeding the infringer or breaching the prohibition of discrimination, and the infringer fulfils its contractual obligations for acts of use already performed in anticipation of the licence to be granted?

2. If abuse of a dominant market position is already to be presumed as a consequence of the infringer’s willingness to negotiate: Does Article 102 of the [Treaty on the Functioning of the European Union] TFEU lay down particular qualitative and/or time requirements in relation to the willingness to negotiate? In particular, can willingness to negotiate be presumed where the patent infringer has merely stated (orally) in a general way that it is prepared to enter into negotiations, or must the infringer already have entered into negotiations by, for example, submitting specific conditions upon which it is prepared to conclude a licensing agreement?

3. If the submission of an acceptable, unconditional offer to conclude a licensing agreement is a prerequisite for abuse of a dominant market position: Does Article 102 TFEU lay down particular qualitative and/or time requirements in relation to that offer? Must the offer contain all the provisions which are normally included in licensing agreements in the field of technology in question? In particular, may the offer be made subject to the condition that the standard-essential patent is actually used and/or is shown to be valid?

4. If the fulfilment of the infringer’s obligations arising from the licence that is to be granted is a prerequisite for the abuse of a dominant market position: Does Article 102 TFEU lay down particular requirements with regard to those acts of fulfilment? Is the infringer particularly required to render an account for past acts of use and/or to pay royalties? May an obligation to pay royalties be discharged, if necessary, by depositing a security?

5. Do the conditions under which the abuse of a dominant position by the proprietor of a standard essential patent is to be presumed apply also to an action on the ground of other claims (for rendering of accounts, recall of products, damages) arising from a patent infringement?

Figure 6. The regulator may come to the rescue of prospective licensees when the SEP licensor’s demands are anti-competitive, too onerous or otherwise fail to satisfy FRAND conditions

© 2014 Archimedes Intellectual Properties, Ltd

Wathelet’s response

Without going into the details of the opinion, we simply present the answers proposed by Wathelet. The full ECJ has not yet opined on this case and the issues that it raises. However, it appears that many of the outstanding issues in SEP/FRAND licensing may be in the process of being resolved:

1. The fact that a holder of a standard-essential patent (SEP) which has given a commitment to a standardisation body to grant third parties a licence on FRAND (Fair, Reasonable and Non-Discriminatory) terms makes a request for corrective measures or brings an action for a prohibitory injunction against an infringer, in accordance with Article 10 and Article 11, respectively, of Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights, which may lead to the exclusion from the markets covered by the standard of the products and services supplied by the infringer of an SEP, constitutes an abuse of its dominant position under Article 102 TFEU where it is shown that the SEP-holder has not honoured its commitment even though the infringer has shown itself to be objectively ready, willing and able to conclude such a licensing agreement.

2. Compliance with that commitment means that, prior to seeking corrective measures or bringing an action for a prohibitory injunction, the SEP-holder, if it is not to be deemed to be abusing its dominant position, must – unless it has been established that the alleged infringer is fully aware of the infringement – alert the alleged infringer to that fact in writing, giving reasons, and specifying the SEP concerned and the manner in which it has been infringed by the infringer. The SEP-holder must, in any event, present to the alleged infringer a written offer of a license on FRAND terms which contains all the terms normally included in a license in the sector in question, in particular the precise amount of the royalty and the way in which that amount is calculated.

3. The infringer must respond to that offer in a diligent and serious manner. If it does not accept the SEP-holder’s offer, it must promptly present to the latter, in writing, a reasonable counter-offer relating to the clauses with which it disagrees. The making of a request for corrective measures or the bringing of an action for a prohibitory injunction does not constitute an abuse of a dominant position if the infringer’s conduct is purely tactical and/or dilatory and/or not serious.

4. If negotiations are not commenced or are unsuccessful, the conduct of the alleged infringer cannot be regarded as dilatory or as not serious if it requests that FRAND terms be fixed either by a court or by an arbitration tribunal. In that event, it is legitimate for the SEP-holder to ask the infringer either to provide a bank guarantee for the payment of royalties or to deposit a provisional sum at the court or arbitration tribunal in respect of its past and future use of the patent.

5. Nor can an infringer’s conduct be regarded as dilatory or as not serious during the negotiations for a FRAND license if it reserves the right, after concluding an agreement for such a license, to challenge before a court or arbitration tribunal the validity of that patent, its supposed use of the teaching of the patent and the essential nature of the SEP in question.

6. The fact that the SEP-holder takes legal action to secure the rendering of accounts does not constitute an abuse of a dominant position. It is for the national court in question to ensure that the measure is reasonable and proportionate.

7. The fact that the SEP-holder brings a claim for damages for past acts of use for the sole purpose of obtaining compensation for previous infringements of its patent does not constitute an abuse of a dominant position.

While we cannot foresee whether the ECJ will take the same position as Wathelet, his answers offer a number of interesting solutions for many of the issues discussed above. In particular, the proposed algorithm would ensure that both hold-up and reverse hold-up issues could be avoided, or at least their impact mitigated. We believe that this approach offers advantages for licensors (in terms of timing of negotiations) and to licensees (protection from injunctions, unless they behave irrationally).

The response requests that the patent owner explain the method it used to calculate the amount of requested royalties. This still leaves the door open for it to justify its request through arguments related to the technical value of its patent(s). Visibility of the calculation algorithm is one important input for the evaluation if a specific offer might be problematic in terms of royalty stacking avoidance.

A possible recourse to arbitration or national courts by the licensor is not explicitly mentioned in the response, but seems to be a rational option, given that the licensee has been given this option explicitly.

Conclusion

SEP/FRAND licensing is serious business involving significant sums of money. Licensees and licensors have sought clarity through the world’s legal systems for various issues related to the rights to use the technology disclosed in essential patents. The courts appear to be in the process of resolving many of the outstanding issues. As a result, the future of SEP licensing is likely to be more standardised and predictable for both licensees and licensors.