Yesterday the European Commission issued its much awaited guidance on the EU approach to Standard Essential Patents. Whilst the guidelines broadly endorse Unwired Planet v Huawei1 and Huawei v ZTE2, what is more notable is what has been omitted.
The Commission has been lobbied hard, in particular by two groups, the Fair Standards Alliance (FSA) and IP Europe. These groups represent either side in the IP debate. The Fair Standards Alliance's strong core of Apple, Intel, Google and Cisco have long lobbied towards the reduction in royalties for Standards Essential Patents, and in the FSA they are accompanied by many of the European automotive manufacturers. IP Europe includes Airbus, Ericsson, Nokia, Qualcomm and Fraunhofer, and has been more supportive of SEPs.
The two key areas of dispute between them were use-based licensing, and where in the chain to license.
Use based licensing
Use-based licensing is the idea that a person who makes a greater use of technology should pay more for it. IP Europe argues that a connected car makes heavy use of wireless technology to drive itself, to navigate, and to entertain its occupants. It should pay more in euros per device than the seller of a vending machine which connects a few times per month to report stock levels. The technology adds more value to the car. The FSA argues for a ban on use-based licensing: if SEP holders can charge different rates depending on the use of the technology, companies who adopt the technology in their products would be required to pay a portion of the value they themselves create.
What lies behind this debate is a fear (or hope) that, if regulators impose a single price, this will inevitably become the lowest common price across all products. The manufacturer of a cellular module for vending machines will argue that cellular connectivity adds only a small value to the machine, that his device sells for only a few euros, that profit margins are slim and that a fair royalty is only a few cents. The heavy users, the car or smartphone makers, who might otherwise pay royalties of euros per device, can drive down their own royalty costs by arguing that they should pay no more than the vending machine module. That might knock a few euros off the price of your next smartphone or car, but, say IP Europe, if SEP holders receive only a few cents for each, no one will develop the next generation of wireless technology.
Where in the chain to license
The FSA also lobbied against the current convention to license at the end user manufacturer level. It argues that all component manufacturers in the chain have a right to a FRAND license. Unless European companies are able to buy and sell fully licensed components, they will lag behind their international competitors. Many companies in the FSA have a valid commercial interest in selling a "licensed" connectivity module: in the automotive industry the model has generally been for the supplier to deal with IP issues, whilst in the telecoms industry SEP licensing has typically been the end user manufacturer's problem.
IP Europe argued that a license to all is a "license to kill innovation in Europe". This is not because it opposes component manufacturers becoming licensed, but because of the effect of laws of exhaustion. Again, behind all of this is an argument about royalty levels, and it's an argument previously seen as the SSPPU approach. A manufacturer of a chip which sells for a few euros can justifiably argue that he can only afford to pay a few cents on each product for wireless connectivity technology. If he can then sell a chip which carries with it the benefit of a license to all wireless technology patents, the smartphone manufacturer, or car manufacturer, can save themselves euros per product in royalty payments. This is why the SEP owners say that the argument about “entitlement for everyone” in the chain is simply a mechanism to drive down royalty payments.
It was widely rumoured that DG Comp supported the FSA's position on the first of these points. At a LeadershIP conference in Brussels in September, Antti Peltomaki, the EC’s Deputy Director General of the Internal Market, proposed that regulators may indeed be able to assist by defining the level of the distribution chain at which SEP licensing should occur.
But the first hints that this might be too controversial came on 10 November when a meeting between various directorates-general that was supposed to finalise the wording of the guidelines broke up without resolution. This was welcome news to those who argue that regulators should be careful in their interventions, particularly now that the courts have shown themselves to be capable of resolving many of the issues that have troubled the industry to date (see our previous update).
It seems that DG Comp has agreed to compromise and the Commission refrained from intervening on these points. As both IP Europe and the FSA have welcomed the lack of any comment, this seems to have been a wise approach.
3 areas of clarification
With the Internet of Things bringing new entrants into the wireless arena, the Commission has stated3 that there are three areas where SEP licensing could benefit from clarification: 1.transparency of information on SEPs 2.debate around the valuation of patented technologies and the definition of FRAND, and 3.uncertainty in enforcement of SEPs.
It set out to clarify these areas in the guidelines:
The Commission argues that SDO databases are not very user friendly, and recommends improving accessibility of data, and eliminating duplications and other flaws. It proposes links from SDO databases to Patent Office databases, so that they may include updates of patent status, ownership and transfer. Although this sounds straight forward, it is actually a considerable exercise and has been the focus of product offerings including Bird & Bird's own patent portfolio intelligence offering Pattern.
The Commission encourages two measures to improve the relevance of declarations, and reduce over-declaration:
•patent declarers should make reference to the section of the standard that is relevant; and
•SDOs may introduce fees for confirming SEP declarations after standard release as an incentive to SEP holders to only maintain relevant declarations.
The Commission suggests that an independent party could implement a new mechanism for “reliable scrutiny” of declarations to assess essentiality. This would be an expensive step, so it would be necessary to "calibrate the depth of scrutiny": in other words this must be relatively high level if it is to be cost effective. Again this is an area that has been addressed to some extent already: reports assessing essentiality of patents declared to cellular standards by PA Consulting are generally considered to be the most thorough and up to date, but other third party reports have been published from time to time.
But the Commission’s call on SDOs, such as ETSI or others, to “urgently” ensure their databases comply with the “main quality features” set out by the Commission suggests that the Commission may be reconsidering whether some standardisation activities fully comply with Article 101. The guidelines might be a warning that we may see formal proceedings based on Art. 101 TFEU involving SDOs in the future. The guidelines contain a disclaimer that they do not bind the DG Comp in any way as regards the application of 101 or 102.
IP valuation and FRAND Licensing
The Commission gave guidance on IP evaluation principles which largely reflect Unwired Planet.
It proposed that licensing terms should bear a clear relationship to the economic value of the patented technology. The Commission recommended that the value should not include any element resulting from the decision to include the technology in the standard. This is perhaps the most controversial statement in the paper and in this area the Commission appears to have departed from the decision of Birss J. in Unwired Planet4.
The issue is as follows. The technology described in an SEP can have three types of value. First, there is the value of the technology per se (i.e. how much would a person pay to use the technology if the standard did not exist). Second, there is the value that the technology contributes as a part of the standard (for example if a manufacturer would pay €1 to use the standard in his products, and there are 10 equally valuable technical contributions that make up the standard, is each worth 10 cents?). Third, there is the hold-up value that an SEP holder could theoretically demand (in the absence of competition law) as an alternative to injunction. In the above example, that might be 99 cents. Implementers tend to argue that the patent owner may only charge the first of these. SEP holders argue for the second. Everyone agrees that the third is abusive.
On initial reading, the commission's guideline appears to favour the first of those valuation approaches, and Judge Posner in Apple v Motorola5 made a similar statement. But the Commission goes on to recognise that this makes little sense where SEPs have no measurable value outside the context of the standard, and in such cases other methods must be considered. These would include the relative importance of the technology compared to other contributions in the standard. That, in the above example, would be the second approach. So we can expect that debate to continue.
The Commission noted that FRAND valuation must ensure continued incentives for SEP holders to contribute their best available technology to the standards. Endorsing Birss J in Unwired Planet it suggests that to avoid royalty stacking, parties need to take into account a reasonable aggregate rate for the standard, that the non-discrimination element of FRAND means right holders cannot discriminate between "similarly situated" implementers, and that a country-by-country licensing approach is inefficient and not be in line with recognised commercial practices.
The Commission also encouraged the creation of patent pools or other licensing platforms within the scope of EU competition rules,6 to increase efficiency of licensing.
Predictable enforcement environment
The Commission endorsed the Huawei v ZTE tests as applied in certain German Higher Regional Court decisions for the availability of injunctions. But it also noted that the availability of injunctive relief is also governed by Article 3(2) of the IP Enforcement Directive7 and the requirement that relief be "effective, proportionate and dissuasive". A proportionality assessment needs to be done carefully on a case by case basis in order to avoid spill over effects of an injunction into other areas, but the Commission notes the necessity for effective and dissuasive relief to alleviate hold-out.
The Commission recognises that SEP holders and licensees may litigate and license on a portfolio basis. It confirms that it would not be abusive for SEP holders to offer more patents, including non-SEPs, but an SEP holder should not require a licensee to accept a licence for these other patents as well. The Commission suggests that portfolio licences would be particularly efficient if they include “all SEPs that a licensee needs”, and the refusal by the implementer to take such a licence may be an indication of bad faith. Whilst this is likely to be understood as a distinction to patent specific licence offers, it does not address licensing strategies of patent pools which the Commission, as mentioned above, is clearly favouring.
ADR such as arbitration is encouraged. Together with the EU IPO it is taking steps to facilitate further use of ADR, in particular by SMEs.
With these guidelines, the Commission considers that the European litigation system (including the Unified Patent Court) include sufficient safeguards to protect against the potentially harmful effects of abusive patent assertion.
Overall, this seems to be a positive step. There is little in the guidelines that either SEP licensors or implementers would strongly object to and the Commission has done a good job in drawing together various sources of guidance in this area into a single document and endorsing the good ones. Most commendably, it has avoided putting too heavy a hand on the scales in favour of either side in the debate.